(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "On The Interpretation Of Some Doubtful Words"

THE BOOK WAS 
DRENCHED 



[<OU_1 60480 >m 

-< 




This 



h'rom Ju Jciwivdl Oj I, ~ 3ihar and Oil^a JRo 
Vol. XVII, 1930-31, Part I. 



II. On the Interpretation of some Doubtful 
Words in the Atharva-Veda 

By Dr. Tarapada Chowdhury, M.A., B.L., Ph.D. 



[The system of transliteration followed here differs from that adopted 
in the Journal in representing ^ by r, ^ by 1, ^ by c, ^T by r, the 
.Ammaeika * by ~ , antflin marking only the UdStta accent (').] 

Prefatory Note 

This is an attempt to find the value of a certain number 
of obscure words found in the Atharvaveda. The suggested 
readings and meanings are by no means absolutely certain ; 
they are to be taken for what they are worth. The main 
criteria (besides context) in determining them have been the 
following: (1) text-criticism, (2) linguistic principles holding in 
the Indo-European or Jndo-Aryan languages or in Sanskrit 
itself, (3) uses, remarks, or parallel passages ill later literature, 
and (4) facts otherwise known. 

In suggesting an emendation of the text there is always a 



Abbreviations 

AB.Aitareya Brahmana : Ait. Aran.=Aitaroya Aranyaka : Ap. Sr.= 
Apastamba rautasutra : ASS.=Anandas'rama Sanskrit Series : AV.= 
Atliarvaveda: Av.=Avestic: Baudh. Sr. =Baudhayana ra\itasutra : 
} >eng.= Bengali : Bloomfield=M. Bloomfield, Hymns of the Atharvaveda, 
Sacred Books of the East, vol. 42 : Boisacq=fi. Boisacq, Dictionaire 
Etyrnologique de la Langue Grecque. BR. =B6htlingk and Roth, Sanskrit 
YVortorbuch : Brugmann=K. Brugmann, Grundriss der vergleichenden 
( Irammatik der indogermanisehen Sprachen : Ch. 8= Chambers, No. 8, 
marmwcript of the Atharvaveda: Dhanva. and Raja.=Dhanvantari- 
and Rajanighantu : Dar.=Darila, commentator of Kaudikasutra : GBr.= 
(lopatha Brahmana : Geiger, Pali Gramm. =Pali Litteratur und Sprache 
by W. Geiger: GOS.=Gaekwad's Oriental Series: Grassmann=H. 
Orassmann, Worterbuoh zum Rigveda: Grill =J. Grill, Hundert Lieder 
des Atharva-veda : Guj.=Gujrati : H.=Hindi: Henry = V. Henry, French 
translation of books vii-xiii of the AV. : Ind. Stud.=Indische Studien : 
Kau^, Kau6ika Sutra: KeS.=Kelava, commentator of Kaudikasutra : 
K8.=Kathaka-Samhita: Ludwig=voL iii of his Der Rigveda: Man. 



26 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

tacit assumption that (1) the mss. either originate from the 
same source or could possibly be compared with and corrected 
by each other, and that (2) the tradition has been for the 
most part a written one. The results speak for themselves : 
but the following facts may be considered. (1) Most of the 
mss. come from Gujarat (12 out of the 20 authorities of 
S. P. Pandit, who, at p. 16 of the introduction to his edition, 
admits giving mostly the Gujarat tradition) ; (2) S. P. Pandit, 
Atharvaveda, vol. Ill, pp. 216 and 315, remarks (on the 
basis of the traditional readings of certain words as compared 
with their developments in the later languages) that the 
tradition of this Veda is derived from mss. and is not strictly 
oral. As to the time when these graphic confusions [viz. 9 of 
p (v) and y fa), kt (m) and tt (IT), etc.] may have occurred, 
Biihler's Palseographic Charts IV, col. XVI and VI, col. V 
show that about the sixth or seventh century A.C. the res- 
pective letters began to be sufficiently similar to be confused 
with each other. 

dksu- 

mrtyor asam fipadyantam ksudham sedim vadhdm bhaydm : 
tndra cd'ksujdlabhyam Sdrvah sendm amum hatam. 8. 8. 18. 

dksum opadm vitatam sahasraksdm vis f uvdti : dvanaddham 
abhikitam brdhmand vi crtamasi. 9. 3. 8. 



Sr. =Manava SrautaSutra : Mar. =Marathi : MBh. =Mahabharata : MS. - 
Maitrayam-San hita : Nigh. =Nighantu : Nir. =Nirukta : Nep. =Nepali : 
Oldenberg=IJgveda, Text critische und exegetische Noten. Pan.=Panini : 
Pet. Lex. or PW. = Sanskrit Worterbuch by Bohtlingk and Roth : Pischel, 
Prak. Grarnm.=Grammatik der Prakrit Sprachen by R. Pischel: Ppp.= 
Paippalada rescension of the AV: PTS.=Pali Text Society: Punj.= 
Punjabi : RV.=?-gveda : ^ankh. Aran.=SahkhSyana Araiiyaka : SB. or 
SBr.=Satapatha ^ -ina : SBE. = Sacred Books of the East : SPP.=, 
Shankar Pandu;| ^*i f fts editor of the Bombay edition of the AV. : 
_' Tand. Br.=Tandya Brahmana: TPr.= 

Taittiriya Prati4akhya:^^=Taittiriya Sathhita : Up. =TJpanisad : v.= 
vorse: Ved. Chrest. = Vedische Chrestomathie : Ved. Concord. =Vedic 
Concordance by M. Bloomfield: Ved. Myth. =Vedic Mythology: Ved. 
IS tud.= Vedische Studien : Vj.=Vaijayanti of Yadavapraka^a, edited by 
G. Oppert: VS.=Vajasaneyi-Samhita: Wack.=J. Wackernagel, 
Altindische Grammatik: Walde=A. Walde, Indogermanisches Worterbuch. 
Wh.= Whitney, Translation of the AV. 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 27 

a vam danaya vavrtiya dasra gor ohena Taugryo nd jivrik : 
apdh ksoni sacate mahina vam jurnd vam dksur dmhaso yajatra. 
RV. 1. 180.5. 

Previous Scholars: The commentary on both the AV. passages is 
wanting. Say an a on the RV. passage takes dkau- as an adjective 
(=vyaptah) qualifying fai- or yajamana- understood; apdh, according 
to him, means either antarikaam or udakani ; ksoni is either nom. du. 
(=dyavapfihivi) or loc. sing. (-=zbhumyam). 'Eine Art Netz', BK. 
' Wagenachse ', Nachtrage to the shorter Pet. Lex. Geldner, Ved. Stud. 
l,136f., equates uksu- with jaladandd- in AV. 8. 8. 5 and renders it 
with 'pole', which in RV. 1. 180. 5 is ' die Stange der Wagens=Deichsel '. 
He is not, however, sure of the epithet sahasraksd- in AV. 8. 8. 18. See 
also his notes, Trans, of RV. vol. I, pp. 233f. Henry (who renders 
with * perche ' in AV. 8. 8. 18) on AV. 9. 3. 8, Bloomfield, SEE 42, pp. 586 
and 598, and Oldenberg, RV. 1. 180. 5, cannot accept Geldner 's meaning 
because of the epithet sahasraksd-, which they understand to mean ' with 
thousand holes ' (Bloomfield). Accordingly Bloomfield, I.e. p. 598, sug- 
gests that a covering of wickerwork ' might be meant in AV. 9. 3. 8. 
This, of course, does not fit in with the other two occurrences of the 
word. Grill renders, 'Flechtennetz'. 

It seems that dksu- is only a bye-form of dksa-, both 
originally meaning 'pole', but the latter eventually being 
restricted to the pole between the wheels of a chariot, namely, 
the axle.. Such duplicate forms ending in -a and -u are not 
rare, cf. for instance : krkavaka- AV. 20. 136. 10 : krkavaku- 
1 hen, 5 kridd-l&V. 1. 37. 1, 5 : kridu- ' playing 5 ib. 9. 20. 7, tanyata-f. 
RV. 1. 180. 12 : tanyatu- < thundering ', turvd&a- : turva&u~, n. pr., 
pada- : padu~J&V. 10. 27. 24 c foot ', mdya- : -mayu- in durmayu- 
RV. 3. 30. 15 ' magic ', aya-L RV. 3. 55. 4 : 6ayu- in Sayutra ib. 
1. 117. 12, 10. 40. 2 * couch ', 6dra- : dru- ' arrow ', harsa- : harsu- 
in harsu-mdnt- RV. 8. 16. 4 ' joy *, etc. 

Prak. akkha- has a meaning * a measure of length, four 
cubits or twenty-four angulis', Seth, Paiyasaddamahan- 
navo s.v., just the same idea as in the English measure of 
ength, ' pole '. Pali akkhavata- * fencing round an arena for 
wrestling' probably contains dksa- in this sense. Beng. akh- 
' sugar cane ', the u at the end of which is still perceptible 
in the pronunciation of Western Bengal, seems more likely 
to have come from dksu-, because of its resemblance to a 



28 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

bamboo pole, than from iksu-. 1 Beng. (Manbhum, dial.) has 
another word alchun to denote a hooked bamboo pole, used 
for separating straw from corn. 

It seems also probable that Or. agtvr) "axe, battle-axe'. 
Lat. ascia (<*acsia, cf. viscus : t6s, vespa<*vepsa), Goth, 
aqizi, old H. Germ, acchus, Germ, axt, old Ice. ex, ox, Ang. 
S. cex, Boisaeq, Greek Dictionary, s.v. agivr] and Eng, axt 
are connected with dlcsu dksa-, the meaning in these cases 
having conceivably been transferred from the 'handle of the 
axe ' to the * axe ' itself. 

We shall now see how this meaning suits the passages- 
quoted above. In AV. 8.8.18 asa- is an uncommon word 
which Whitney, followed by Henry, emends to osa- anc 
translates mrtyor asam apadyantdm as ' let them go untc 
death's burning (?)*. He remarks, however, that only twc 
of his MSS. read osam, all the rest asam, ' which must according 
ly be regarded as the traditional text, although unintelligible ' 
It is probably the same word as dsd- ' nearness ' cerebralised b\ 
mistake, helped probably by the nearness of the preceding 
r. a Mrtyor asam, etc., can then be paralleled by 3. 11. 2b=RV 
10. 161. 2b, yddi ksitayur yddi vd pdreto yddi mrtyor antikdw 
mtd evd. If this be acceptable, we may modify Whitney'* 
translation thus : " Let them go unto the nearness of death 
unto hunger, debility, the deadly weapon, fear; by pole anc 
net, O Sarva, [do thou] and Indra slay yonder army ". Poles oi 
the net or the club are referred to as instruments of killing 
in the same hymn : v. 5, anidriksam jalam asid jaladand? 
dUo mahih : tend 'bhidhaya ddsyundm Sakrdh sendm dpdvapat , 
v. 11, trnidhv endn matydm Bhavdsya ; and v. 12, Sddhya ekm\ 
jaladanddm udydtya yanty ojasd: Rudra Skam Vdsava ekan 
Adityair ka udyatah. 



is given as a synonym of iksu-, Dhanvantari- and Raja- 
nighantu, ASS., p. 156. Hindi ukh (besides ikh) is probably a case oi 
assimilation, iksuh> ikkhu>* ukkhu> ukh. Cf. Pr. ucchu^iksu-, Brug- 
mann, Kurze verg. Gram. d. idg. Sprachen 330. 

* Such irregular cerebralisation is found elsewhere in the text. Cf., 
for instance, ftubhyaa tva 3. 10. 10, sdth nas \ibhib 2. 35. 2, vi tasthirc 
4. 6. 2, etc. 



VOL. XVII. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 29 

AV. 9. 3. enumerates the different things used in the 
construction of a Sola, a particular kind of thatched house 
still in common use in India. A description of a modern &ala 
with reference to the Atharvan hymn may be found helpful in 
properly understanding the hymn as a whole and the verse 
under discussion in particular. In the MSnbhum district, 
before the walls are raised, a frame-work of the house is made 
by fixing posts (upamit- 1 . v. 1) at short distances which are 
connected at the top by posts running all round (parimit-, v. 1). 
On these connecting poles rests one end of the slanting cross- 
poles (8amdam6d- 2 , v. 5), which, with their other ends, hold the 
dividing pole at the top (pdrisvarijalya-* , v. 5, visuvdnt-, v. 8). 

These are laid over with sliced pieces of bamboo (prob. 
paladd-*, v. 5 and 17) tied to the crosspoles by means of 
strings, made of a kind of gmss, or bamboo-skins (v. 4). The 
top of this frame- work is then covered with straw, which is 
kept together by means of sliced bamboos spread upon it and 
tied to those below (v. 17). Walls are then raised, or, in some 
cases, the supporting posts are first connected by a net-work of 
rushes (v. 18), which is then plastered with earth. Sometimes 
the crosspoles of the thatch hang down a little beyond the walls 
and are then supported by a series of slanting poles, fixed 
against (pmtimit-, v. 1) the outer sides of the supporting poles. 
Of furniture in such a house, two pieces are almost invariably 



1 Upamit-, parimit-, and pratimit- are all from the root mi- * to build *, 
' to erect'. Upamit- is found also in RV. 1. 59. 1, 4. 5. 1, where a support- 
ing pillar or post is intended, as well as mit- 9 ib. 10. 18. 12, with the same 
meaning. 

2 SamdaMd- usually means * tongs '. These poles are so called, 
because each pair of them holds the ridge-pole as a pair of tongs. In 
Beng. they are called kaci ( scissors ', which gives the same idea. 

3 Pdrisvaftjalya- is probably a corruption forsvanjanya-, * that which 
is to be embraced (viz., the ridge-pole, by the cross-poles on either side) ' ; 
cf. Ppp. reading, parisvaficanadasya. For interchange of n and I, cf. 
kanydna- RV. 8. 36. 5 : kanydla-AV. maiden ', pttani-L : pefala- ' beauti- 
ful', mulali-AV : mpnali- lotus-stem ', and Wack. 1. 175 (c). n. 

* Paladd- seems to be a Prakritiem for pra-rada-, containing the root, 
rod- in the sense of * scraping', * slicing', 'splitting', cf. tdaara- : traaara-* 
'shuttle'. 



30 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

to be found : the sling (6ikyd-, v. 6), made of ropes, for hanging 
vessels on and the bamboo pole, sliced or intact, for hanging 
clothes on. The latter is hung down either from the crosspoles 
or from the ridge by means of cords tied to either end. When 
heavily laden, it would, of course, make a curve and thus 
resemble a head-band (opa6d-, for which see below). This must 
be intended by dksu- in the verse under discussion (viz., 9. 3. 
8) ; the epithet ' thousand-eyed (sahasraksd-) ' which, under 
any circumstances, would be an exaggeration referring to 
its knotty joints. These, oval and slowly rising, resemble the 
eyeball to some extent, and for this reason they are often 
referred to as the * eyes ' of bamboo, cane, sugar cane, etc., in 
Bengali and other modern Indian languages. 1 

Accordingly, we may translate: "The out-stretched 
thousand-eyed pole, (like) a headband, hung down from (dva~ 
naddha-) and tied on to (abhihita-) the ridge, do we unfasten 
with our spell ". The two epithets dvanaddha- and abhihita- 
are significant as referring to the clothes -pole. 

The RV. verse may be translated : " May I turn towards 
(i.e., be inclined to give) a present for you, O wonderful ones, by 
the offering of a cow, like the aged son of Tugra (viz., Bhujyu). 
Through your greatness, from water he goes to the earth ; by 
you two (vam) his pole of distress is shattered, O adorable 
ones." 

The pole here intended must be that, with slings at both 
ends, used for carrying loads (vivadha- or vihangika-, Hindi 
bahang), the idea being that he was thereby relieved of the 
load of distress that he had been carrying. I take here apdh as 
ablative singular of dp-, cf. Grassmann, s.v. dp, for use in 
singular; ksoni as instrumental singular of ksom- 'earth', 
which is quite in consonance with the pada-patha's considering 
it a pragrhya, cf . Ved. Gr. 72. 2 (c), and the pada-text of astri 
AV. 6. 27. 3b, mahi 18. 1. 39b etc. ; mahina=mahi(m)na 9 the first 
vowel lengthened for the sake of metre and then confused with 
mahina-, which accounts for the accent ; cf . pdthistham AV. 

1 Nepali has akhafo for notches, which seems to be an extension of 



VOL. XVH. PT. I.J INTERPRETATION OF DOUBTFUL WORDS. 31 

14. 2. 6d (pathes tham, RV. 10. 40. 13d), ' accented as if it were a 
superlative' (Wh.); and vdm in d as instrumental dual, cf. RV. 
1. 158. 3a, yukto ha ydd vdm Taugryaya perur vi mddhye drnaso 
dhayipajrdh, and ib. v. 4d, ma mam edho dd6ataya& cito dhdk prd 
ydd vdm baddhds tmdni Tchadati ksam. For the story of Bhujyu's 
deliverance from the waters onto the dry land, cf. RV. 1. 116. 
3-5, 117. 14, etc. 

dksya- 

ehi jtvdm trayamdnam pdrvatsya 'sydksyam : viSvebhir devair 
dattdm paridhir jivandya kdm. 4. 9. 1. 

Previous Scholars : * parvatasya trikakunnamno girefy aksam asi cdksur 

bhavasi' Sayana. ' bist eino Salbo vom Gebirge ( ? ),' Grill, who 

referring to 19. 45. 3 (parvatiyam afijanam), suggests an emendation to 
afijana-. Quoting Roth's opinion that dksya- may mean ' belonging to 
the eye, i.e., an eye-ointment ', he remarks that the characteristics of the 
remedy do not admit of such a limitation and that he can rather conceive 
of an epithet aksayya- or aks ara-. " Come hither ! th ju art the living, pro- 
tecting eye-ointment of the mountain given by all the gods as a safeguard 
unto life."- Bloomfield, who remarks in the notes, ** Aksham does not 
mean * eye ', akshyam is otherwise unquotable. Nevertheless we have trans- 
lated akshyam, for the passage seems to be a tantalising reverberation of 
Sat. Br. III. 1. 3. 12 (viz., yatra va Indro Vftramahams tasya yad aksy aslt 
tarn giriih trai-kakydatn akarot tad yad traikakydam bhavati caksusy evai 
'ta$ caksur dadhati). He refers also to MS. III. 6.3. and TS. VI. 1. 1. 5, each 
of which gives a version of the same story. " Come thou, rescuing the 

living one ; of the mountain art thou for the eyes (?) " Whitney, who 

remarks, " The meter indicates that the true reading at the end of b is 
aksyam....; but dksya is unknown elsewhere and its meaning in this 
connection is obscure." 

The bad construction and obscurity of meaning which follow 
from the usual way of splitting b seems to suggest that we have 
to deal here with two words instead of three, viz., parvatasya 
dsydksyam, ' living in the mouth of the mountain '. This would 
be a fitting epithet for anjana-, one variety of the soft rock from 
which it is made being found in the cavities of some mountains, 
the other in the beds of certain rivers ; cf. v. 10, yddivd'si traika- 

kuddm yddi ydmundm ucydse : ubhe te bhadre namni , and 

aauviranjana- and sroto'njana- in Vaidyaka, e.g., Dhanva. and 
Raja., p. 125, Bhavaprako-.sa, Srlvehkatesvara Press ed., p. 43. 
Asydksya- (to be read dsiaksia-) contains the present stem of the 



32 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

root ks i- ' to live ' (of. pres. ksiyanti) followed by the suffix -a, 
formed like cakram-asajd- RV. 5. 34. 6, a-jpa/-d- ib. 1. 148. 5, 
sadaprn-d- ib. 5. 44. 12, etc., Ved Gr. 115. 3a, which (denoting 
the agent) normally accent the suffix. The verse may thu* be 
translated: "Come thou! rescuing the living one, who <trt a 
liver in the mouth of the mountain ; given by all the gods, an 
enclosure, as it were, for life." 

The confusion of the copyist of the pada-patha is easily 
accounted for in the reminiscence of the story referred to above 
and in the fact that anjana- generally means an eye-ointment. 
I say, 'generally', because there are indications in the same hymn 
and elsewhere that it was applied to other parts of the body as 
well, being thus equivalent to an ordinary ointment or medicated 
oil. Cf. v. 4., ydsya'njanaprasdrpctfydngam-angampdrusparuh : 
tdto ydksmam vi badhasa ugro madhyamaSir iva ; v. 8, trdyo 
dasa anjanasya takma balasa ad dhih ; and 19. 45. 5, aksvai y kam 
manim ekam krnusva snahy ekena pibai 'kam esam : cdturviram 
nairtebhya& caturbkyo grahya bandhebhyah pdri patv asman. 

In this connection may be considered dhi- in v. 8 quoted 
above. The present state of accent and the padapatha require 
us to take adaJiih as two words and the commentator accord- 
ingly explains dhi- as ' snake-poisoning.' But coming together 
with takr 'n- (fever) and balasa- (cold ?) l it would rather refer to 
some malady as Whitney also has suggested. No disease of 
the name, however, is known. I should think that it is one 
word, addhi- 'burning, i.e., the sore caused by burning,' cf. 
ajani- * birth' RV. 3. 17. 3, abhogi- ' enjoyment' ib. 1. 113. 5, 
samtani- 'clashing noise', etc. As ointments, including medi- 
cated oils, are prescribed for fever, cold, and burnt-sores, in 
their various stages, this meaning would suit the context quite 
well. 

ddruksna- 

ydt te vasah paridhanam yam ntvim krnuse tvdm: Sivdm 
te tanve tat krnmah samsparSe 'druksnam astu te. 8. 2. 16. 



is used frequently in the medical Sastras as equivalent 
to kapha- or tlesman- ' phlegm ', ' expectoration.' Cf. Astangahrdaya, Sutra. 
23, 19 ; Uttara. 40, 31 ; and B.B., s.v. 



VOL. XVII. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 33 

Previous Scholars : ' tac ca vaatraih aaihsparte viaaye cidrukanam 
arukaam yatha mardavam Canute (for aatu te) vyapnoti gacchati tatha 
kfnmab.' SSyana. '. .nicht rauh sei's zur bertirung dir ' Ludwig, p. 497. 
' . .puisse ce contact t'6tre une caresse ' Henry. * . .not rough to thy. 
touch a) "t be ' Bloomfield. ' . . be it not harsh to thy touch ' Whit- 
ney, noting that " SPP. reads ' with all his authorities ' ddrukanam in d. 
Our mss. might doubtless ail be understood in the same way, but some of 
them look more like -du- or -du- ; -rw-, which our text unfortunately 
gives, is not found in any; neither ruksna nor drukana appears to be 
met with elsewhere." 

Adruksna- is undoubtedly the correct reading, which even 
the corrupt 'duksanam of Ppp. points to. The word apparent- 
ly contains the root druh- ' to harm', followed by the suffix -sna, 
Ved. Or. 118, and preceded by the negative particle. The 
absence of aspiration in -dru- points to the fact that -ks- in 
cases like this represents an earlier -gzh-, 1 which has been lost in 
Sanskrit, Wack. I. 209. a. 

The lengthening of the root-vowel is similar to that found 
in tikmri- 'sharp': tlj- 'to sharpen', tigmd-, TS. hdKksna- : 
VS. kaliksna- 'a kind of animal', stksa- : Siksa- 'phonetics', 
which, according to Wack. I. 39. n., have their I from the 
desiderative, where I is found several times before -ks- } partly in 
place of older i. To these may be added niksana- : niks- 
' pierce/ and probably diks- ( to initiate a pupil ' : di& ' to 
direct' and ruksa- 'rough': ruj- 'to break', 'to harm', with 
the same underlying idea as in ddruksna- (BR. connect it 
with TUS- 'to make dusty', which is not convincing, both as 
regards meaning and change of -s- to -k-, cf. Wack. I. 118. n.). 
As instances of the lengthening of u gives Wack., loc. til., 
pratyavaruhya : pratyava-mh- (Aufrecht, AB. 427), niruhya, 
AB. 7.5.1: nir-uhya, tusnim 'silent': tus- 'to be silent', VS. 
sutflnd- : elsewhere, sumnd-, Up. suksma- : VS. SB. suksmd-. 

It will be seen in the above instances that the i or u 



1 Other instances from the RV. are : ddksat 1.130.8, 2.4.7, dak*i 2.1. io, 
daksi (voc.) 1.141.8, dakrisah (g.s.) 1.141.7, dtikaoh (g.s.) 2.4.4 : dcth- 
to burn' ; odukaat 1.33.10, adukaala 1.160.3, dukadh, 7.4.7, dukatin 1.121.8, 
dudukaan 10.61.10, 74.4, dudukaat 7.18.4 : dith- ' to milk ', etc. In all these 
cases the pada-text replaces d (the second one in reduplicated stems) 
by dh after the manner of later Sanskrit. Cf. Wack. I. 160. 



34 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

in each case is followed by a conjunct whose last member or the 
last two, when it consists of three, arc continuatives. Such 
conjuncts are, of course, easier to be pronounced as one whole, 
whether at the beginning or in the middle of a word. There 
grew up thus a dialectical tendency, as testified by TPr. 21.7.9 
(mentioned in Wack. I. 240. b), to consider such a conjunct 
in the middle of a word as belonging to the following vowel. 
The heaviness of the immediately preceding vowel, which 
is usual before a conjunct, being thereby shaken, there was 
occasion for a compensatory lengthening which is what we have 
in the above instances. The reason, then, why this pheno- 
menon is not universal may possibly lie in the fact that the 
spelling was constantly corrected, where etymological connec- 
tions were obvious. 

Adruksna- would thus mean * unharming ' (cf. druhila- 
in Man. $r. 2.14.14,. .druhilam ahatam vasah paridhaya. . , 
* putting on rough, now clothes ') in the above passage : " That 
which is thy cloth for putting round, the waist-strip that 
thou makest ; we make it auspicious for thy body, be it unharm- 
ing in touch to thee." 

ddroghdvild 

krnuld dhumdrn vrsanah sakhayo 'droghavita vacam dccha : 
ay dm Agnih prtanas&t siiviro yena deva dsahania ddsyun. 
11.1.2=RV. 3.29.9, which has krnota. .vrsanam in a, 'sredhanta 
itana vfijam dccha in b, and devaso in d. 

V. I. : Wh/a collation-book gives -avitdh in.fo as padu- reading without 
noto of variant. SPP. gives -avita, following one or two of his nisH. 
and the commentator. Ppp. is corrupt, but has in b, adrogha vita vatam 
matsa. 

Previous Scholars : * adrohakarinam sucaritrfinam avita rakzita vacam 
ace ha . ..fgrupam vacam abhilaksya,' Say. ' Adj. Wahrhaftigkeit liebend ' 
BR. Bloomfield, p. 611, thinks that the Saunakiya text scarcely yields 
sense in b, and that the RV. and the Ppp. texts suggest the reading, 
ddrogha vita vajam dccha or adrogha dveta etc., on the basis of which 
he translates: 'unharmed by wiles go ye into the contest.' Henry 
translates b, * dans la direction de la parole (sacree) confiants dans la 
faveur du (Dieu) inoffensif.' "Make ye smoke, O ye bulls, companions, 
ye that are aided by the unhateful (?), unto speech: the Agni (is) fight - 
overpowering, having good heroes, by whom the gods overpowered 
the barbarians " Wh. 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 35 

Adroghavita vacam dccha seems to be an exact paraphrase of 
dsredhanta itana vajam dccha in RV. Adroghah and dsredhantak 
both mean ' unharming ' or ' unharmed.' For double Sandhi 
between ddroghah and avita, see under eru. The meaning 
'to move', * to go', for av-, given in the Dhatupatha and 
quoted by BR., has to be assumed for several cases : for 
instance, avatkd- in ado ydd avadhavaty avatkdm ddhi pdrvataf : 
tdt te krnomi bhesajdm subhesajam ydtha 'sasi, AV. 2.3.1, 
c what runs down yonder, gliding off the mountain', etc. 1 ; 
dvani-,* ' stream ', ' river ', or ' course of river ', connected by 
some with av- 'to protect' and by others with dva 'down 5 , 
both unsatisfactory : iid-dvantau in ahdm enav udatisthipam 
gavau &rdntasddav iva : kurkurav iva kujantav uddvanlau vfkav 
iva, AV. 7.95(100). 2. C I have made these two stand up 
like two weary-sitting cows: barking like dogs (du.) springing 
up like wolves (du.)' 3 ; pra-av- in marudbhih prdcywta meghah 
pravantu prihimm dnu AV. 4.15.9 c, d, ' let the clouds dropped 
forth by the Maruts glide along the earth ' 4 , etc. For the older 
form avita instead of avata, cf. RV. 7.59.6, a ca no barhik sddala 
*trita ca sparhani dfitave vdsu. 

Vacam is a case of de-voicing, which, so common in the 
Ppp., is also found several times in the Saunakiya text; cf., 
for instance, dva-tlryatih (mss.) for -dlrya- (Wh.) in 19.9.8d, 
6dm no bhumir vepyamdna 6dm ulka nirhatam ca ydt : $dm 
gavo lohitakslrah sdm bhumir dvatlryaiilt : vitavati for vidha- 



1 Whitney renders "What runs down yonder, aiding (?) off the 
mountain," and remarks, " Avatkd (ava tkdm : quoted in the comment to 
Prat. i. 103 : ii. 38 ; iv. 25) is obscure, but it is here translated as from the 
present participle of root av (like ejatkd, v. 23.7. [cf. abhimadayatkd, 
QB., viksinatkd, VS.]) ; this the comm. favours (vyadhipariharena rak- 
salcam)." 

2 Johansson, IF. 2, 62, derives dvani- and avatkd- (m. 2) from au 
I.-B. root eye- to be wet.' Cf. also Persson, ib. 35, 200. 

3 " like (two) growling dogs, like (two) lurking (? ud-av) wolves,"-- 

Whitney, noting, "The comm. explains uddvantau by goyuthamadhye 
vatsan udgfhya gacchantau : Henry renders, that watch one another ' 
[He would reject ud in a] ". 

* Whitney renders c, d : " let the clouds, started forward by the 
Maruts, show favour (pra-av) along the earth." 



36 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.3. 

(BE.) in 12.2.38d (of which c, d are repeated as c f d of v. 
52), muhur grdhyaih prd vadaly artim mdrtyo nitya : kravyad yan 
agnir antikad anuvidvan vitavati ; priyasam for bhriyasam 
(Say., Wh., Weber) in 3.5.4c f Somasya parndh adha ugrdm 
agann Indrena datto Vdrunena fristdh : tarn priyasam bahu 
rocamano dtrghayutvaya Satd&aradaya ; bharcikah (majority of 
SPP.'s and one of Wh.'s mss.) for -rjikah in 18.1.30C ; mathavycin 
for madhavyan in 2.35. 2c ; ydti for yddi (Wh.) in 10.3.6b, 
svapndm suptva yddi pdtyasi papdm mrgdh srtim ydti dhavdd 
djustnm: pariksavac chakuneh papavadad aydm manir varano 

varayisyate, etc. 

(iparaparandh 

avastum enam dsvagam aprajasain karoty aparapamno 
bhavati ksiyute. 12.5.45. (12.11.7). ydevdm vidvso brahmandsya 
ksatriyo gam adattt. v. 40. 

Previous Scholars: *Adj. ohno Forsetzung, ohne Nachkommens- 
chaff, BB. " Elle lo prive de fortune, de patrimoiue, de posterity : il 
n'a plus ni anc^tres ni descendants, il perit. ...,'' Henry. "Celui (viz., 
sense) de aparaparano parait ressortir a la fois du contexte et de Tanalyse 
du mot," ibid., commentary, p. 259. "Without abode, without home, 
without progeny, she rnakos him ; he becomes without succession (?) ; he 
is destroyed : " Whitney. 

BR. and Henry evidently connect aparaparanah with -pa fa- 
para- which expresses the idea of relativity such as, * far and 
near ', { prior and posterior ', c before and behind 5 , etc. But how 
they explain the suffix -na is not clear. In its absence it would 
rather have meant without anything else to stand by, far or 
near, before or after, in future or in the past', ( all alone by him- 
self ', i stranded in the world '. A change in accentuation would, 
however, enable us to see in it two words (with double Saridhi), 
*dparah devoid of riches (: rai- ' riches J ) ' and *dparanah * devoid 
of pleasure ', cf. dpavrata-, dpodaka -etc. For the correspoild- 
ence, a- (before consonants) : %-(bef ore vowels), cf. Wack. I. 91> 
Ved. Gr. 362. Compare also Md-ra- RV. 10. 106. 5, which, 
according to Grassmann and Macdonell (loc. cit.), has rai- for its 
second member. 

araiaki- 

e'yam agann osadhmam virudham vlryavati : aja6rngy arataki 
ttksnarngi vy fsatu. 4. 37. 6. 



VOL. XVII. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 37 

Previous Scholars : " ara adataro himsakah tan asmat sthanat atayati 
uccatayatVti arataki ; t'lksno^^gl tiksne ugragandhe fyhgatytl phale yasyah 

evathgunavtiista" Sayana. " N. der Pflanze Agaorngi oder Beiw. 

derselben Vielleicht verwandt mit arala." BR. " Hither hath 

come this mighty one (virytivant) of the herbs, of the plants ; let the goat- 
horned arataki, the sharp-horned, push out ". Wh. Ludwig and Bloom- 
field also do not translate arataki, nor aja&iftgi, and they consider ftknna- 
dfftgi as if in the instrumental (' with its sharp horns '). 

Arata- in arataki is evidently the same word as arada- l in 
MS. 2. 5. 9, p. 59. 17, arunds tupardt caitreyo devanam asm (?) tyeto 
'yaMrngah ^aineyo 'suranam, te 'sura utkrodino 'carann, arado 
'smakam tupa 3 ro 'misam iti, and aradi- in S.Br. 4. 5. 5. 5, atha 
ygfl upamSy'b, hutva urdhvam amimqrsti tasmad ima aj& araditara 

akrgmamana iva yanti, and in TS, 5. 6. 21. 1, vdruni krsne 

ua6e aradyau 2 divyav rsabhau parimarau. Sayana explains 
aradyau in TS. as ucchrita&rhgaii ' high-horned ', but in view of 
the passage from MS. quoted above, where arada- is contrasted 
with tupard-' hornless', the meaning seems to be simply * horned '. 
The first element in arata-, viz., am-, is most probably connected 
with am- ' spokes ', ala- 'sting of the scorpion (or, of an insect in 
general, cf. alin- ' bee ')', ara-' awl \ etc., all denoting something 
pointed, considering how frequently the idea of piercing is asso- 
ciated with Srnga- l horn ', cf . Mite 6fhge rdksase vinikse RV. 5. 
2.9d, ' he sharpens his (two) horns in order to pierce the demon.' 
The second element is the suffix -/a, so frequently found in 
Apabhraiiisa. It is probably identical with the suffix -to, 8 cere- 
bralised through the influence of the neighbouring r-sound, 



1 Sans, t, between vowels, regularly becomes -rf- in Prakrit, and most 
of the modern Indo-Aryan vernaculars, cf. Pischel, Prak. Gramm. 192 ; 
Geiger, Pali Gramm., 38. 

2 Sayana (followed by others) connects ara4yati with fsabhau and 
accordingly takes ara$yd- as the stem; but in view of the other two pas- 
sages it seems preferable to take aradi- as the stem, the word qualifying 



8 Suffix -ta is found in dn-ap-ta- RV. 0. 16. 3 'not watery*: op- water ', 
dnttia- AV.6.4.2, 8.5.11 near at hand': dnti- near ', dmanyuta- AV. 12. 3. 31 
4 not wrathful ': manyti- wrath ', avatd- < well ': dva down ', pdrvata- 
mountain ' (lit.,* rugged ') : pdrvan-)omt, Mmata- :* troman- (Grassmann). 
etc. It appears as >ita in puapita, phalita, etc. 



38 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

although not immediately preceding. 1 For lengthening of the 
stem -vowel next preceding the suffix and a similar cerebra- 
lisation, cf. krkdta- ' neck- joint ': krka- ' throat ', ^rngdta- ' Trapa 
bispinosa 9 : &rhga~ 'horn'. 

Later Sanskrit ardla- ' bent ', ' curved ', ' crooked ', must 
be a further development of the same word. From 'horned' 
to ' bent like a horn ' seems to be an easy step. Intervocalic 
t > d > 1 (1) is a common phenomenon in middle Indian, cf. 
PLschel, Prak. Grammar, 238 ; Geiger, Pali Grammar, 386. 

As is clear from the quotation from Sayana above and 
from the synonyms visdnikd- and mesa&rngt- given in the 
Dhanvantarinighantti, p. 23, the names and various epithets of 
this plant (Odina pinnata) are due to the hornlike shape of its 
fruit . The last two padas of the above verse may be translated 
"Let the horned aja&rhgi ('goat-horned'), the sharp-horned, 
(or, with its sharp horns), pierce (the Gandharvas and Apsarases, 
mentioned above).' 5 Gf. ardyydm brahmanaspate tikana&rngo 
'drsann ihi KV. 10. 155. 2, &rngabhydm rdksa rsaty dvartim hanti 
cdksusa AV. 9. 4. 17, etc. 

alaji- 

visalpdsya vidradhdsya vatikardsya valajeh : ydksmandm 
.sdrvesam visdm tiiravocam ahdm tvdt. 9. 8. 20. 

Previous Scholars : Alaji- has been rightly identified by BB. and 
others with tdajl- of the medical ^astras, but the meaning of the latter 
has unfortunately been narrowed down, doubtless through an oversight, 
to ' a disease of the eyo\ and Bloomfield goes so far as to render it with 
* inflammation of the eyes '. Henry reads balaji- instead. Whitney does 
not translate it. 

As a matter of fact, alaji- appears to be a kind of boil 
which may be formed in any part of the body. BR. were 
misled by the occurrence of the word among diseases of the eye 
in Susruta, Uttara. Chap. 2, and thought they found support 
in the word andhalaji- (ghanam avaktrdm pldakam unnatdm 
parimandaldm : andhdlajvm alpapuydm tdm vidyat kaphava- 
tajdm, Susr., Nidana. Ch. 13, 6) which is simply a boil without a 



1 Cf. Pa. pati : prati, Pa. pathama-: prathama-, Sarah, darvaghata- 
claas. c.arvaghata-: aghata-, garudti-: garutmant-, class, asrpata- * bleeding' 
(lex.) : rt^-pS/o-, Wack. I. 140 (a), para 5. 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 39 

mouth. Bhelasamhita, p. 91, considers the fifth layer of skin 
(of which it counts six layers) as the base of alaji- and vidradhi-, 
and Susruta, Sutra. Ch. 2. 7, 9, considers it as due to the 
morbidity of the flesh, along with various kinds of tumours and 
swellings (adhimamsa-'rvuda-'ro- 9 dhijihvo-'pajihvo- 9 paku6a-gala- 
Sundika-'laji- ma^msasamghatau- 9 sthaprako2)a-gala[ganda]' ganda- 
mala-prabhrtayo mamsadosajah). Vagbhata counts it among 
boils due to Gonorrhoea (prameha), dahati tvacam utthane bhr6am 
kasta visarpini : raktakrsnatitrt-sphota-dfflia'moha-jvaralaji, 
Astafigahrdaya, Nidana. 10, 39 (cf. Susruta, Nid. 6.14, 18), 'alaji 
burns the skin, while rising, is very troublesome, expansive, 
reddish black, and attended with severe thirst, eruption, inflam- 
mation, fainting, and high temperature '; among diseases of the 
face, gandalaji sthirah Sopho gande ddhajvaranvitah, ib. Uttara. 
21. 12, ' a gandalaji is a firm tumour on the cheek, attended with 
inflammation and high temperature ' ; among diseases of the 
eye, kamnasya'ntar alaji Gopho ruk-toda-dahavan, ib. Uttara. 
10. 9 (cf. Susruta, Uttara 2. 8.), 'alaji is a, tumour in the pupil 
with pain, pricking sensation, and inflammation '; and elsewhere. 
A comparison of the above passages would point to its being a 
boil or tumour similar to vidradhi (abscess), but much smaller. 

dlpa&ayti- 

ye ma krodhdyanti lapita hastinam maSdka iva : tan ahfim 
man ye durhitan jane dtpasayun iva. 4.36.9. 

V. L SSyana has lipitah (=upadigdhah 9 sathkrantah) in a, and 
durhatan (duatahananeria visayltytari) in c. The pada- mss. read lapitn, 
which SPP. emends to -tah. Whitney suggests emendation to lapitvn. 

Previous Scholars : *. . . . jane janasamghe tatsamcarasthale avastithan 
alpaayun, parimanataji, alpakayah fayanasvabhavah samcaraksamah kita 
alpcuSayavah, te yatha pranisamcarena hanyante tadvad aham anayasena 
apunarbhavaih hanmVtyarthahS Sayana. 'Ein bestimmtes l&stiges 
Insekt oder dgl.' -PW. (V. 1056), and others have followed (< Ungeziefer' 
Grill, 'small vermin' Bloomfield, 'mites (?)' Whitney), except Lud- 
wig, who translates : ' die mich erz&rnen, zum sprechen gebracht (wie 
fliegen den elefanten) | mein ich, sind sie ungliicklich, nur kurze zeit im 
volk verweilend.jj' Grill ('durch Gesumm ') and Bloomfield (* with their 
jabber') seem to consider lapita as if in the instrumental. 

The redundant syllable in the first pada and the two obscure 
words lapita (or lipitah) and dlpaSayu-, which can hardly be 



40 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

reconciled with the context in a satisfactory manner, make it 
obvious that the passage is corrupt. The meaning of dlpa&ayu- 
as given by Sayana and hesitatingly followed by others is a 
mere guess. The word does not appear to be found anywhere 
else. &ayu- is found elsewhere in the sense of ' the sleepy one ', 
especially the snake called ajagara, or of ' couch ' as in Sayutrii 
(RV.). Neither of these meanings would suit alpa6ayu-, which, 
as a Bahuvrlhi, would mean ' one having few 6ayus '. On the 
other hand, the adjective durhitan, which means ' miserable ' as 
in RV. 8. 19- 26 (nd tva raslyd 'bhiSastaye vaso nd papatvaya 
santya: nd me stota 'mafiva nd durhitah syad Ague na papaya), 
suggests that dlpa&ayun is probably a corruption for dlpapasuu 
(so also Grill, p. 139). Compare, AV. 12. 4. 25, anapatydm 
dlpapa6um va&a krnoti purusam, and 4. 17. 6, ksudkamardm 
trsnamardm agotam anapatydtam, where, as in many other places, 
want of cattle is considered equivalent to misery. The process 
might have been this : by an unconscious metathesis in the mind 
of the copyist dlpa6apun was written in place of *pa&un, which 
then was easily simplified to ~ayun, helped by the similarity of 
the letters/) (v) and y (*l) 1 as well as by the comparative 
intelligibility of ~&ayu- as against -&apu-. 

The first hemistich seems to have originally read, ye uie 
dhdyanti lohita (or 4dm) Jiastino maSdka iva. The first step seems 
to have been probably the intrusion of the root krudh- from the 
following verse, after which lohita (or -tarn) might have been 

1 Another curious instance of confusion of these two letters, followed by 
a syncopation of the preceding vowel, appears to be abhteastya- in 6. 120. 2, 
bhumir matti 'ditir no janitram bhr&la 'ntdriksamabhttastya nah t : dyaurnah 
pitapitryac chdih bhavati jainim gtvti m& ' va patsi loktit, which is of such an 
obscuring character that Whitney is forced to remark ; * the variants are of 
the kind that seem to show that the text was unintelligible to the text- 
makers, and that we are excusable in finding the text extremely obscure '. 
If, however, we see abhia8tya(h) < -dastiyah < tiastipah, the meaning is 
quite clear : " The earth our mother, Aditi our birthplace, the atmosphere 
our brother, are our protectors against imprecation; may heaven our 
father be weal to us from paternal (guilt) ; having harmed my relatives, 
may I not fall down from the wished -f or-world ". The reading in a, 
dbhitasta bnah, in TA. 2. 6. 29 is due to a different kind of confusion, that 
between p ( ^ ) and e ( ^ ) ; thus, dbhisasta tnah < -tastipandfK -tastipa 
noh. For other examples, see eru. 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 41 

consciously changed to lapita ( ' jabbering ' ? ) in order somehow 
to agree with it and, of course, the genetives me and hastino to 
the corresponding accusatives. Sucking of blood being what is 
expected of the pi&acas, the eaters of raw flesh (cf . AV. 5. 29. 9, 
kravyadam Ague rudhirdm pi6acdm), against whom the whole 
hymn is directed, the verse may be translated (with these emen- 
dations) : " Those who suck my blood as mosquitoes that of the 
elephant ; them I consider wretched, as, among the people, those 

who have few cattle". 

dvarjuslnam 

turanam dturdndm vi&am dvarjusmdm : samaitu vi&vdto 
bhdgo antarhastdm krtdm mama. 7. 50(52). 2. 

tvdm no vdvav esam dpurvyah somdndm prat-hamdh pitim 
arhasi sutandm pltnn arhasi : ido vihunmatindm victim vavarju- 
slndm: viSva it te dhmdvo dnhra a&iram ghrtdm duhrata a&'imm. 
RV. 1. 134. 6. 

Previous Scholars: * dyutakriyam a parity ajantinam ', Sayana 
on AV. "Ob reich siej sind oder nicht, die Leute helf kein Widwrstand ; 

" Grill. ** Of the quick, of the slow, of the people that cannot 

avoid it ( ? ), let the fortune come together from all sides, my winnings in 
hand". Whitney, "The meaning of dvarjwlnam in b is extremely 
problematical; the translators " wehrloss " etc. Comparison with vMm 
vavarju8inani RV. 1. 134. 6. and the irregularity of the unreduplicated 

form, make the reading very suspicious; Ppp. has instead devayatim " 

ibid.y notes. Grassmann and the RV. translators in general consider 
vavarjusmam as containing vtf- ' to spread ', thus meaning ' those who 
have spread the ku^a-grass '. 

The meaniag given to vavarjuslndm in RV. may suit the 
context quite well, but viSam vavarjusinam and vi&am dvarjusi- 
ndm are so much alike that one cannot help thinking that they 
probably express the same idiom and contain essentially the same 
words. Both vavarjusinam ( : vrj-' ' to spread ' ) and dvarjusmdm 
'( ' VJ- 'to give up ' ) are irregular if they are really perfect parti- 
ciples with -vas, the former because there is scarcely another 
example in Sanskrit of -vas with the strong stem of the 
perfect, 1 and the latter because of the want of reduplication. 
The Ppp. reading devayati(nd)m ' worshipping the gods ' in the 

1 Cf. Ved. Gr. 491-92: Brugmann, Grundriss 136. para. 4, "die 
Wurzel silbe erscheint meist in Tiefstufengestalt, regelmassig im Aris- 
chen, ..." The latter, loc. cit. 9 notes some exceptions from Gr. and Goth. 



42 INTBRPBETATION OF DOUBTFUL WORDS. [j.B.O.R.S, 

corresponding passage seems to be a paraphrase of dvarjusinam 
in the Saunaklya text. For the phrase, cf. RV. 1. 36. 1 (vi&am 
devayatinam), 1. 77. 3, 3. 6. 3, 7. 69. 3 (devaydnfir vMh). 
Now, (ivarjusinam, if containing dvas- ' help, protection (of the 
gods ) J and root jus- c to enjoy ', e to relish ', with suffix -i, would 
give almost the same idea. The Rigvedic passage might in 
that case be considered as containing visam u avarjustnam, 
which would first became vi6am v avar-, as in VS. 23. 44, 6dm v 
astu tanve Mva, or AV. 6. 56.3, sdm v asna 'ha asydm, and later 
on, when the meaning had become obscure with the conse- 
quence that vavarjusinam was considered one word, -m would 
naturally be changed to Anusvara. Change of -as to -ar is not 
regular, but there are instances. Cf. dnar-vi&-. RV. 1. 121. 7: 
anas- and Wack. I. 28 (y). Or else, which is more likely, 
uvar- may be a parallel stem with dvas-, just as amnar-: amnas- 
(Pan. 8. 2. 20). vsdr-: usds-, udhar-: udhan-: udhas-; dhar-: 
dhan-; ratharyati : rdtha-, vddkar- : vadhd-, vanar- in vanargu-, 
vandrsdd-: vdna-, saparyati-: sap- ' to worship ', etc. 

Avar- in avir mahd Indra dadrhi 6rudM nnli 6u66ca hi dyauh 

fcsa nd bhlsa adrivo ghrnannd bhisa adrivah RV. 1. 133. 6, 

in the hymn just preceding the one considered above and attri- 
buted to the same Parucchepa Rsi, may also be the same word. 
It has been so far considered an instance of avds- f downwards ', 
the only instance of its ending in r. But its identification 
with dvas- gives a much better meaning, when dadrhi (which 
has been tabulated by Grasumann under dar-, dir-, c to burst ', 
f to split up ', with a query) is, at the same time, considered a 
form of the root dr- * to take notice of, 'to regard ', which 
would be quite regular : " Take notice of our great cordial 
(Grassmann, s.v. dvas-, meaning ii), O Indra, ( and ) listen to 
us : for the heaven has burnt like the earth (?) for fear, fo* 
fear of heat, O you carrying stone . . . . " If that be the true 
meaning, the explanation of the present accentuation lies in the 
confusion of dadrhi as above noticed. 

aSvaksabha- 

isirn yosa yuvafir ddmuna ralri devdsya Savitur Bhdgasya : 
asvafaabhfi w-hdva sdmbhrtaMr a paprau dyavaprthivi mahitva, 
19.49.1. 



VOL. XVII. PT. I.] INTEBPBETATION OF DOUBTFUL WOBDS. 43 

Previous Scholars : " aauni a6uni svavisaye 4lghra-pravfttlni aksani 
caksuradmdriyani abhibhavati tiraskarotVti crfvaksabha, cak*uradinirodhike 

*li yavat yad va 'atvasya budhnam purusasya mayum J ity uttaratra 

vakayamanatvad ayam arthah : advan ksapayati kmpayatl 'ti aAvaksa, 
atvaksa bhd dlptir yasyah so." Say an a, " The lively woman, household 
maiden, night, of god Savitar, of Bhaga, all-expanded, of easy invocation, of 
assembled fortune ( ?-pn), hath filled heaven -and-earth with greatness ". 
Whitney. " In c, all the mss. , with the comm. and SPP. , read a$vak#abhd 
(aqvd ksabhti), which, as being unintelligible, our edition emends at a 
venture to vigva-vyacas, and the translation follows the latter, for lack of 
anything better Ppp- reads erpva/tarra." Ibid., notes. 

The correct reading may probably be a6vaksubhfi ' restless 
oi- swift like a horse ' : ksubh- ' to put in motion/ * make 
restless/ cf. ksubha (Grassmann, " ksubh, f., schiielle Beweg- 
ung") in RV, 5. 41. 13, vida tin nu mahanto ye va eva brdvanto 
dasma vary am dddhanah : vdya& cand subhva a'va yanti ksubha 
mdrtam dnuyatam vadhasnaih. If this be the true reading 
and meaning, it would lend the idea of swiftness to the verbs n 
paprau in d, and dti aruhat in v. 2 a. 

asamsuklagild- 

Rudrdsyai 'labakarebhyo ' samsuktagilebhyah : iddm mahv- 
syebhyah Svdbhyo akaram ndmah. 11. 2. 30. 

Previous Scholars : To Rudra's howlmaking, unhymned-swallow- 
ing (?), greatmouthed dogs I have said this homage/' Whitney, who 
notes, "The obscure asamsuktagild (Ppp. -girebhyas) is paraphrased 
by the comm. with asamicmam o^obhamana-vacanarh gjrnaUi bhasante. 
How aaamsukta should come to mean ' unmasticated ' as given in the 
Pet. Lexx. does not appear. The translation given above conjectures ' not 
having a hymn with it ' ". " Aux hurleurs de Rudra, | qui devorent ceux 
qui n'ont point d'hymnes [a leur chanter]. . . ., " Henry, who remarks in 
the commentary, " Je Us asulcta * depourvu d'hymnes,' le sens implicite 
4tant: *par consequent, ils ne nous deVorent pas, nous qui avons un 
h^mne a te chanter ' ". 

The correct reading appears to be asamsulta-giUbhyab 
' swallowing what is not properly cooked (or not cooked at all)' ; 
sutta-, past participle of svad- (alternating with sud~, cf. suda- 
'cook' and Wack. I. 20 and 79 b) Ho make tasteful', 
'to cook', formed in the same way as nuttd-: nud-, vittd-: 
vid-, aattd-: sad-, etc. The mistake may be due to one of 
the following reasons: (i) tt (IT) and kt (n) are very easily 



44 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

confused: (ii) kt>tt in Pali, Prak. and the vernaculars, this 
being known to the reader or writer, a mistaken attempt to 
re-Sanskritise it is conceivable ; and (iii) aukta- being in much 
more general use than sutta-, an emendation of the latter, 
if it is not understood, to the former is also conceivable. It 
will be seen that ' uncooked- swallowing ' is a much happier 
epithet for dogs than any other that could come from asam- 
sukla-. 

dharjata- 

sdm vo gosthtna susdda sdm rayya, sdm subhutyd : dhar- 
jdtasya ydn nama Una vah sdm srjamasi. 3. 14. 1. 

a tva crtalv Aryama Pusa Bfhaspdtih : dharjdtasya ytin 
nama Una tva 'ti crtdmasi. 5. 28. 12. 

Previous Scholars : ' ahany ahani jayata ity aharjatah pranivtiesah, 
tasya yan nama aharjata iti tena namna.' Say ana on 3.14.1. Whitney 
renders dharjatasya ydn nama in both the places with 'that which ia 
the name of the day born one' and comments (under 3.14.1), "The 
obscure third pada is found again below as V. 28. 12.C; it is altogether 
diversely rendered (conjecturaliy) by the translators (Weber, with the 
blessing of favourable birth"; Ludwig, "with all that which one calls 
day-born"; Grill, "with whatever a day of luck brings forth"); R. 
suggests 'with all (of good things) that the day brings, or that is 
under heaven ': none of these suits the oth^r occurrence." " c aus- 
picious ' comes very near its sense. Its opposite is anahar-grata, ahkh. 
6'r. XIV, 51, 2-5 ' born on an unlucky day ' =papa-nakaatre gratah, Kaus, 

46, 26 and elsewhere Either it is <born on a good (puraya) day' 

or ' born by day in distinction from night ' (cf. naktamgrata I. 23. 1.) 

with the name ' might mean with kind or species', cf. V, 4, 8." Bloom- 
field, p. 351. 

It seems that dharjata- here probably refers to the god 
Bhaga and that there is a pun upon the word, as it also means 
'luck'. 1 Cf. AV. 19.45.9, Bhdgo ma bhdgena'vatu ; 3.16.5, 
Bhdgo evd bhdgavS astu devds tena vaydm bhdgavantah syama ; 
14.1.34, sdm BMgena sdm Aryamna sdm Dhata srjatu vdrcasd, 
etc. Bhaga is frequently mentioned together with Aryaman, 
Pusan, Brhaspati, etc. (cf. AV. 3.20.3, 6.4.2, 6.74.1, 14.1.50, etc.), 

1 Cf. Macdonell, Ved. Myth., p. 45, " The word bhaga also occurs 
about twenty times in the RV. with the sense pf ' bounty, wealth, fortune ', 
and the ambiguity is sometimes played upon." 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 45 

and might naturally be expected in 5.28.12 as well as in 3.14.1 (the 
latter immediately followed by, sdm vah srjatv Aryama sdm Pftsa 
sdm Brhaspdtih : sdm tndro yo dhananjaydh ....). In RV. 1.123.5, 
Usas is called sister of Bhaga, which suggests that the latter 
was a morning deity, probably a particular aspect of the sun. * 
AV. 3.16. (=BV. 7.41.), devoted to Bhaga, is composed as if to 
be chanted in the morning. This supports the above contention 
and makes it plausible that pratarjitam in v. 2, which is 
generally translated as ' early-rising ' and which Sayana (on RV.) 
takes as two separate words, is probably a corruption of 
prntarjdtdm e born in the morning ', essentially the same as our 
aharjala-. 

dndika- 

esd yajnandm vitato bdhistho vistdrmam paktva divdm a 
riveSa : dndikam kumudam sdmtanoti bisam Sdlukam 6dphako 
inuldli: etas tvd dhard upa yantu sdrvdk svarge loke mddhumat 
l*invamdnd upa tvd tisthantu puskarinih sdmantdh. 4.34.5. 

na'sya kse.lre puskarim na'ndikam jay ate blsam : yasmin 
rdstre ninidhyHe brahmajdya 'cityd. 5.17.16. 

Previous Scholars : " This extended, is of sacrifices the best carrier ; 
having cooked the vistarin, one has entered the sky; the bulb-bearing 
lotus spreads (sain~tan), the bisa, Qal&ka, gdphaka, midali : let all the 
streams (dhfira) come unto thee swelling honoyedly in the heavenly 
(wxirga) world; let complete (sdmanta) lotus-ponds approach thee." 
Whitney (4.34.5). "Not in his field is a lofcus-pond, the bulb ( Ibisa) of 
the bulb-bearing lotus is not produced (jan), in whose etc., etc." ibid. 
(5.17.16). Under 4.34.5 he notes, " The mss. (with the exception, doubt- 
lees accidental, of our P.K.) all read bdhisthas at the end of a, and this 
SPP. retains, while our text makes the obviously called for emendation to 
vah-. The things mentioned in c, d appear to be edible parts of water- 
lilies ; the bulbous roots, leaf -stems, and radical fibres, which in some 
species, as the Nymphnea esculenta, are savory, and which are eaten some- 
what like asparagus. . . .The kumuda is the N. esculenta (kairava, comm.) ; 
and the comm. explains bisa (he reads vi$a) as the root-bulb of the padwa 



1 Cf. Ved. Myth., p. 44, " In the aggregate sense they (i.e., the 
Adityas) are the gods of celestial light, without representing any particular 
manifestation of that light, such as sun, moon, stars, or dawn ", and p. 45 
" Dawn is Bhaga's sister (1.123.6). Bhaga's eye is adorned with rays 
(1.136.2), and hymns rise up to Visnu as on Bhaga's path (3.54.14). Yaeka 
describes Bhaga as presiding over the forenoon (Nir. 12.13) ". 



46 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

(Nelumbiumapeciodum) [of. Lanman, JAOS. XIX. 2d. half, p. 151f.], $aluka 
as that of utpala (a Nymphaea), $aphaka as a hooi-(capha) shaped water- 
plant, and mulall &9~mfnzli. Qaphaka occurs also at Ap. (}&. IX. 14.14., 
where it seems to signify an edible plant or fruit, perhaps a water-nut ". 

Of the six names given in 4.34.5c, d, the meanings of 
three are definitely known : kumuda-ia the water-lily, 6aluka-i& 
its root, and mulali is the root of the lotus (Pali has exactly the 
the same form), of which the latter two are edible. Bisa-, in 
later literature, is often confused with mrndla-(mulaU-) ; it is, 
however, distinct from the latter (cf. mrnalam ca bisanvitam, 
Caraka, Cikitsa. 11, 78 ; bisani ca mrnalam ca. . , . , ibid. v. 82 ; 

bisa-mrndla-ka6eruka-srngataka , Susruta, Sutra. Ch. 21.) 

and must originally have meant the stem of plants like water- 
lily, lotus, etc. The water-lily grows a bulbous (egg-shaped) 
container of seods and tho lotus, one which looks like a horse's 
hoof with a circular flat top gradually tapering downwards until 
it joins the stem; both of these are edible (i.e., only the seeds). 
The former * may be the dndika- and the latter the aphaka-*. 
The three pairs of words in the above verse stand in such a way 
that one in each pair seems to be the subject of sdm tanoti and 
the other its object, thus : " The water-lily spreads the ' egg- 
like ' (at the top), the (lily)-stem the lily-root (down), (and) 
the lotus-root the * hoof -like ' (at the top) ". This would imply 
emendation of dphako to -kam. A similar emendation seems to 
be required also for vitato and vdhislho (as emended by Wh.) in 
a, which would then agree with vistarinam and thus leave esdh 
free to be the subject of paktva and a vive&a. 

a&uhgd- 

nir balase 'tdh prd patd 'Aungah 6isuk6 yathd : dtho ita 
iva Kay and 9 pa drahy dmraHd. 6.14.3. 

Previous Scholars : ' he baldsa yathd yena prakdrena atuftgah 

afiugami 6u4ukdfy etatsamjflo m^go duram dhdvati tadvat gaccha\ Sayana. 

1 It is interesting to note that it is called bh&t in Bengali, while what 
is known as * egg -fruit ' in English is called bhata in Hindi. 

2 tfaphaka in Ap. r. 9. 14. 14, kfsnajinafii kuto vd karno va gardabho 
harino vd harinapfndkd va tydmaka-pdtro vd saphako ve'ti vijftdyate, is 
explained by the commentator as * a deer that has lost a hoof * and 
Caland follows him in his translation. 



VOL. XVH. FT. I.] INTERPRETATION OF DOUBTFUL WORDS. 47 

"M. wohl so v. als ciSuga, N.eines Thieres, vielleicht eines Vogels 

Moglichist die Auffassung (in AV.): wie ein Fiillen, das zum Rosse(a#tt) 
Wft." BR. " Fly forth from here, O balasa, as a swift foal (after the 
mare). And even, as the reed in every year, pass away without slaying 
men ". Bloomfield. " Fly forth, O balasa, like a young OQurhgd, " Whit- 
ney. 

In the present reading it is doubtful whether we have to 
take d&ungdh or SiSukdk as the substantive. If following BR. 
and Whitney we take the former as a noun, denoting some bird, 
nnd&i&ukdh as its adjective, the comparison falls flat, since a young 
bird can hardly fly. On the other hand, if aSumgd- is an adjec- 
tive, the formation causes difficulty. As a rule, when, in a 
compound word, a nominal stem has a nasal just before the 
verbal root following it, it is simply the sign of the accusative 
of that stem. A6u- being an adverb can only have a&u in the 
accusative. The only other apparent exception to this rule is 
maksumgamd- in maksumgamabhir utibhih RV.8.22.16 [cf. Wack. 
2. 1. 86 (e). n : Ved. Gr. . 276 (p. 164. fn. 5.) ]. But maksumga- 
md- may be explained as meaning ' going to the quick ( in sacri- 
fice)', not * quickly-going '. 

8i6u(kd)- means 'the young of an animal or man* and it 
often stands for the calf as in RV. 2.34.8, dhenur nd 6i6ve : the 
latter may be its meaning also here, its running away (at the 
approach of man, owing to nervousness), which is the point of 
comparison, being well-known. In view of this -td ' Suhgdh may 
be considered a corruption for -ta a$rngdh, helped by the pro- 
nunciation of r as ru in some parts of Southern India, cf. tabhi 
sluptOySrupto, or stuto ior-bhis-trpto in AV. 19.4.1d (Wh.'s notes). 
Afyhgdh ' hornless ' would imply that the calf is very young and 
would thereby support the point of comparison. 

dsravd- 

ydthd dyam ca prthivim cd'ntds tisthati tAjanam : evfi rogam 
cd 'srdvdm cd 'ntds tisthatu munja it. 1. 2. 4. 

ad anga kuvid anga atdm ya bhesajani te : tesdm asi tvdm 
uttamdm andsrdvdm droganam 2. 3. 2. mcaih kkananty dsura 
arusranam iddm mahdt : tad dsrdvdsya bhesajdm tad u rogam 
amnatiat. v. 3. (of which c and d are repeated as c and d of vs. 
4 and 5 ). 



48 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

satdm ya bhesajani te sahdsram sdmgatani ca: Artstham 
asrava-bhesajdm vdsistham roganadanam. 6. 44, 2. 

Previous Scholars : Say ana explains asravd- as mutrati8ara-( diabetes) 
in 1.2.4, atisara-'timutra-nadwranadayah (diarrhoaa, diabetes, ulcer, etc.) 
in 2.3.2, and raktasrava- (bleeding) in 6.44.2. "Ein Korperschaden, 
Gebrechen", BR. * Diarrhoea', Bloomfield and Grill. The latter re- 
marks (p.80) : " 2.b. dsrdvd nicht : Fluss, eitemde Wunde,. . . .eondern. .. 
Durchfall (Darila : atisdra). Die dreimalige Zusammen-stellung mit rdga 
lasst fur letztres die Bedeutung: Leibweh (cf. ruj f.) annehmen (so 1st z.B, 
Qiroroga=$ir$akti Kopfweh bei Dar)." " Asravd is rendered by the indefi- 
nite term * flux ', its specific meaning being uncertain ; " Whitney, 
under 2.3.2. 

It seems to me that the specific meaning * bleeding ' is 
required in all the above occurrences of asravd. The following 
are the reasons : (1) AV. 1. 2. is about the defiance (v. 2) and 
avoidance (v. 3) of the arrows of the enemy. V. 4. would fit in 
with the other verses only if it relates to a wound already made 
and its cure ; otherwise, as Whitney suggests, it would seem 
unconnected. See Kaus. ; 14. 7 and 12, and Whitney's intro- 
duction to the hymn. (2) Hymn 2. 3. contains the word arusrana- 
c wound-healer ' twice (vs. 2 and 5), and appears to be directed 
only against the healing of wounds and stoppage of bleeding. 
(3) Eoga-j found in every case with asravd-, seems to have a 
special meaning here, probably a derivative and archaic Oiie, 
'wound', lit. ' breach (in the body)' :rnj- 'to break'. This 
would help to clear the obscurity of the second half of the first 
verse quoted above, showing at the same time its special connec- 
tion with asrdvd-, and would also give a point to 6. 44. 1 : " The 
heaven has stopped, the earth has stopped, this whole world has 
stopped : the erect sleeping trees have stopped, may this thy 
wound (i.e., its bleeding) stop." And lastly (4), 6. 44. 3 contains 
the word vatikrta-na^anl which as shown by Zimmer, Altindis- 
ches Leben, p. 389, means healer of the wound '. Thus all the 
three hymns in which asravd- is found are solely directed to- 
wards the healing of wounds. l 

4 Vatlkfta- is found again at 6. 109. 3 together with ksiptd- 
(' bruised', Wh.) and atividdhd- ('pierced'), and vatiktfra- at 9.8(13).20 
together with visalpd-, vidradhd- and alaji-, all referring to an abcess or 
boil of some kind. Zimmer identifies vata- with Germ. ' Wunde '. It 



VOL. XVn. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 49 

uttaradrau 

mesa iva vai sdm ca vi co'rv dcyase ydd uttaradrav uparas 
ca khadatah : Sirsna siro 'psasa 9 pso arddyann am&un babhasti 
hdritebhir asdbhifr. 6. 49. 2. 

Previous Scholars : Sayana, following Kau6. 46. 14, considers all the 
three verses in the hymn as addressed to Agni by the pupil on the occasion 
of his preceptor's cremation. His reading is slightly different, having 
ucyase uc samavaye) for acyase in a, aparah for uparah in b and 
bibhasti(=bhak8ayati) in d. He explains uttaradrau either as upary- 
avasthita-kaatha-yukte dahyaSarire or utkrstataradrume mahav^ksabhuyia^he 
vane and supplies Agnih as subject of babhasti in d. " wie ein beutel 
[ blasbalg ] ziehst du zusammen weit dann auszeinander dich, wenn der 
oben laufende [ uttara-drah ? ] stein und untere f reszen, | drangend kopf 
an kopf, wange an wange friszt er die stangel mit grunem maul." 
Ludwig, p. 432. " Like a ram, thou art bent together and wide apart, 
when in the upper wood the upper and the lower stone devour ; exciting 
(ard) head with head, breast (dpsas) with breast, he gnaws the soma- 
stalks with groen mouths." Whitney. 

The verse is one of three forming a hymn, which seem to 
have been originally unconnected (see Whitney, introduction to 
the hymn). As it stands it hardly gives any connected sense, 
and the text (given by Wh. in the notes) as found in Ppp., 
KS., or Ap. Sr. is still worse. The correct reading seems to be : 
mesa iva vai sdm ca vi co 'rv dcyase ydd uttara dvav uparas 
ca khadathah : slrsna siro 'psasa 'pso arddyann amsun babhatsi 
hdritebhir dsdbhih. " Like a ram, thou art bent together and 
wide apart, when, O upper stone, thou and the lower one, the 
two, devour; pressing head with head, breast with breast, 
thou gnawest the soma-stalks with green mouths." 

As Whitney also has suggested alternatively, the action of 
the stones here is compared to that of the rams, butting and 
drawing back, in a ram-fight, a favourite pastime even now in 
India. The corruption would be explained by a graphic confu- 
sion in the change from dva- (^T) to dra- (5fT), followed by an 
emendation of khadathah to -tah, and a metathesis in babhatsi 

seems to contain a root van- ' to cut ' which is probably to be traced also 
in van- vana- ( wood ', vamtid- ' bamboo ' (cf. iqksa- ' tree ' : vratic- ' to cut ') 
v&i- *axe' etc. If this is identical with van- * to like', 'to enjoy', the 
development of meaning may be traced thus : * to cut'>* to divide '> to 
share *>' to enjoy '. 



50 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

to -ati, the former being the second person, singular, present of 
root bhos- t viz., ^babhos-si (for change of s to t, cf. Wack. I. 
153). 

Similar must be the case with babhasti in v. 1 of the same 
hymn, ?iahi te Ague tanvah krurdm andmsa mdrtyah: kapir 
babhasti t&janam svdm jarayu gaur iva, which Whitney trans- 
lates, " Surely no mortal, Agni, hath attained the cruelty 
of thy self (tanu). The ape gnaws (bhas) the shaft (ttjana) 
as a cow her own after-birth." Here the second half-verse 
as it stands, presents no traceable connection with the first, and 
as nothing is known about the ape's gnawing a shaft, the com- 
parison is obscure. But an emendation of babhasti to -tsi and 
t&janam to te jdnam not only brings out a good sense, but also 
supports Kausika's injunctions : " The tawny one, thou gnawest 
thine own man (i.e., worshipper, namely, my preceptor), like 
a cow her after-birth." For kapi- in the sense of f tawny ', cf. 
Abhidhana-Rajendra, s.v. kavi, and also kapila- and kapiAa-. 

rdantu 

tndra jahi pumamsam yatudhanam utd striyam maydya 
xasadanam : vignvaso muradeva rdantu ma te drsant suryam 
wcdrantam. 8. 4. 24.=RV. 7. 104. 24. 

Previous Scholars : rdantuna6yantu, sfiJadanam^himsantMn. 
S5y an a, RV. and AV. BR., followed by Grassmami, consider it a form of 
v/ewd- and give tho moaning 'in Bewegung (der Theile) gerathen, 
zerstieben, sich auflosen '. " O Indra, frappo le sorcier male | et la femelle 
qui s'enorguil!6t de sa magic : ]] que le demons au col tors soient an6antis : 
| qu'ils ne voient pas le soleil se lever." Henry. "O Indra, smite the 
man sorcerer, likewise the woman who is prevailing with magic (mayfi) ; 
let the neckless false- worshippers vanish (gd) ; let them not see the sun 
moving upwards." Whitney, with the remark, ** The obscure rdantu is 
glossed by the comm. with napyantu". 

Besides here the root rd- is found in the following typical 
instances: (a) ardrd- RV. 1. 116. 4, 2. 13. 6, AV. 1. 32. 8, 
and in the compounds danu- AV. 16. 3. 4, VS. 18. 45, pavi- 
AV. 16. 3. 4, pavitra- AV. 9. 6. 27, Q hasta- AV. 12. 3. 13; 
(b) rdu- in rduddra-, epithet of Rudra RV. 2. 33. 5, of Mitra, 
Varuna, and the Adityas ib. 3. 54. 10, of Soma ib. 8. 48. 10, and 
rdupi (p. rdu* pt) and rduvfdha (p. rdu* vfdha) RV. 8. 77. 11 ; 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 51 

(c) ardan dhdnvani RV. 4. 17. 2 ; (d) kdtdm ardati AV. 12. 4. 3 ; 
(e) ardati ' to pray/ ardana- ' prayer,' later Sans. ; (/) prardayo 
mcir apdsah samudrdm RV. 6. 17. 12; (g) udadhim arddya 
AV. 4. 15. 6, ardayati ib. v. 11 ; (h) pdrdncam Susmam ardaya 
AV. 6. 65. 1, (i) vrtrdm arddy- RV. 1. 187. 1, 10. 142. 2, Ayum 

Kutsam, Atithigvdm 8.53.2 vi parirapah 2.23.14; (j ) 

Sirsna &iro 'psasa 'pso arddyan AV. 6. 49. 2 and later Sans. 

Outside Sanskrit it is found in the Gr. verb apScu, dpSeuco 

* to sprinkle, to water ' and in the derivatives, dp8/*o? ' water 
for sprinkling, 5 d/>8aviov ' vase of water ' ; also in Av. awdvt 
k goddess of the waters.' Boisacq, Greek Dictionary, s.v. apSco. 

All facts considered, the original meaning appears to bo, 
(1) * to be or make wet or watery/ from which, through 
association of ideas, must have developed the following mean- 
ings: (2) from the impression of running water or rainfall, 
(i) ' to glide/ (ii) * to fall ' ; (3) from moistening one's heart 
through prayer, ' to pray, to ask for ' ; (4) in the causatives, 
(i) c to make something, such as water, glide/ (ii) ' to make fall/ 
'to send down rain/ etc., (iii) * to make an enemy fall/ i.e., 

to defeat or kill ' ; (5) from 4 (iii) ' to press or oppress." l 

We shall now see how these meanings fit in with the above 
typical instances. 

To begin with, (a) ardrd- not only means ' wet ' (i.e., 
soaked with water) as in RV. 2. 13. 6, yd bhojanam ca ddyase ca 
vdrdhanam drdrad a suskam mddhumad dudohitha, but also 
4 watery ' (i.e., composed wholly of water) as in RV. 1. 116. 4, 
tisrdh ksdpas trir aha ' tivrajadbhir Nasatya Bhujyum uhathuh 
patangaih : samudrdsya dhdnvann drdrdsya part tribhi rdthaik 



1 Prof, bivsg thinks there are probably two different roots, (1) fd- 
1 to water ' contained in the Gr. and Av. instances, in ardrd-, and probably 
in fdu-f and (2) fd- * to fall ' contained in the other instances. But con- 
sidering the relationship of dru* * to melt,' and dru~ ' to glide,' * to run 
fast,' of gal- to melt ' and gal- * to slip or fall,' H. girna * to fall,' and 
of Eng. verbs drip and drop, it seems not unlikely that they are one 
and the same root in different stages of development of meanings. Walde, 
s.v. erd> " (zer)fliessen, Feuchtigheit," also connects ardati, fddti * flows, 
etc.', ardayati * kills, etc.', ardrd- 'wet,' fduh- 'moisture', and Av. 
awdvi-, although he separates Gr. apSw etc., from this group. 



52 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

6atdpadbhih sdlavaih, and in ardrddanu- 'having watery drops/ 
cf. VS. 18, 45, samudro 'si ndbhasvan ardrddanuh. 

(b) Yaska, Nir. 6. 8, equates rduddra- with mrdudara- 
and is followed by Say ana and BR., the latter giving the 
meaning 'mild, sanft, gnadig.' Following the traditional 
accentuation Grassmann splits up the word into rduddra- 
(suggested also by BR.) ; the pada-text does not divide it. As 
there is hardly another instance of loss of an initial m, Grassmann 
is evidently right in holding that an element rdu- 1 (which has 
nothing to do with mrdu-) is to be seen in the three words rdu- 
ddra-, rdupe, and rduvfdha. Rduddra- probably has the 
same meaning as vapodara- in RV. 8. 17. 8, Iwvigrivo vapodarah 
subahur dndhaso made : tndro vrtrUni jighnate, which does not 
appear to mean ' obese ' as taken by Grassmann, but rather 
'one having fat (here a special variety of it, namely, ghee) 
in his stomach ' (like vajajathara-, epithet of gharmd-, 5. 19. 4) ; 
cf. the epithets, ghrtannau RV. 6. 67. 8, ghrtasutl 1. 136. 1, 
2. 41. 6, and ghrtdyoni 5. 62. 2 of Mitra-Vdruna, ghrtasuti 
6. 69. 6 of tndra-Visnu, and ghrtaSri 10. 65. 2, ghrfdsnu- 9. 88. 5, 
of Soma. Compare also vapavant-, epithet of Agni RV. 6. 1. 3, 
with ghrtdvantam yonim of Agni RV. 10. 91. 4. Thus rdu-= 
vapa- or ghrtd- ( : rd with meaning 2. j.), and the second element 
is uddra-, not ddra-. 

This meaning of rdu- suits also RV. 8. 77. 11, tuviksdm 
te suhrtam sumdyam dhdnuh sadhur bundo hiranydyah : ubha 
te bahu rdnya susamskrta rdupe cid rduvfdha, of which the fourth 
pada seems to have presented a difficulty ever since the time of 
Yaska. Sayana simply quotes Yaska, Nir. 6. 8, e esa nirukte 
ekam api padam vihaya Yaskena vyakhyata, tad eva likhyate. 
tuviksam bahuviksepam mahaviksepam te sukrta r h sumayam 
susukham dhanuh sadhayita te bundo hiranyanyah: ubha te 
bahu ranyau ramamyau samgrdmyau va 9 rdupe ardana-patinau 
marmany ardanavedhinau va. 9 How the form rdupe (not, 
however, considered a pragrhya in the pada) can qualify, as 
an adjective, a masculine noun bahu does not appear. BR. and 
Grassmann take it to mean ' Biene oder ein anderes siissigkeit 
suchendes Thier, f .' ; but even then the force of cid ' even ' 

i For this word, see also IF 2, 27. 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 53 

is lost and we have to assume it here as equivalent to iva. 
It is probably the locative singular of rdupd-, ' in the sucking of 
butter,' used in a verbal sense like avaghrah (=avaghranam 
smelling ') in Ap. Sr. 8. 16. 3, 12 ; 13. 17. 9 ; 15. 11. 14. The 
adj. susamskrta- ( well-polished ' seems also to imply the rubbing 
of Indra's arms with ghee (or some other oily substance), which 
would afterwards shine and thus give out all the more the gloss 
due to it (rduvfdh-). Thus we may translate : " Many-killing, 
well-made, well-finished is thy bow, magnificent, golden is (thy) 
arrow: both thy arms are fit for the fight, (rdnya- <rdna- 
1 fight '), well-polished, which, even in sucking butter, are 
butter-increasing." 

(c) Ardan dhdnvani in RV. 4. 17. 2, tdva tviso jdniman 
rejata dyaur ejad bhumir bhiydsd svdsya manyoh : rghdyanta 
subhvdh pdrvatasa ardan dhdnvani sardyanta apah, probably 
means ' the dry lands became wet ' (I), 1 cf. the contrast between 
the wet sea and the dry land in RV. 1. 116. 4, quoted above, 
and dhdnvant srotah krnute ib. 1. 95. lOa or ire 'va dhdnvan 
ni jajdsa te visdm AV. 5. 13. 1. 

(d) Katdm ardati in AV. 12.4. 3, kiitdya 'sya sdm 6iryante 
slondya katdm ardati: banddya dahyante grhah kandya dlyate 
svdm, definitely means ' falls into a pit '. 

(e) Ardati in the sense of ' requesting, praying (3) ' is found 

only in classical Sanskrit, as in Raghuvamsa 5. 17, 

nirgalitambugarbham Saradghanam nd 'rdati cdtako 'pi. 

(/) In the instance quoted above it means c thou madest 
(the waters) glide forward (4. i.) '. 

1 Prof. Sieg thinks that here too, as in (d), the root r,d- means ' to 
fall ', the imagery, according to him, being that the mountains dropped 
down stones on to the dry land and water below, and made the latter 
move. He, together with several other authorities, considers that the 
passage refers to the violent movements on the earth's surface during 
a volcanic eruption or earthquake. This, however, is not sufficiently 
clear from the context, which seems rather to indulge in a glorification of 
ludra by stating that fixed states of things alter when he is angry, cf 
in particular v. 13. Another objection is that the construction of the 
verse is such as to imply that dhdnvani and tipah are as much subjects 
to the verbs attached to them as dyauh, bhumih and pdrvatasah to those 
attached to them. 



54 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.8. 

(g) In the AV. instances quoted above it means * send down 
(the cloud) (4, ii) '. 

(h) Ava manyur dva 'yata 'va bahu manoyuja: pdra&ara 
tvdm tisam pdrancam Susmam ardaya 'dha no rayim a krdhi, AV. 
6. 65. 1, is translated by Whitney : " Down (dva) be the fury, 
down the drawn [arrow], down the mind-yoked arms. 
demolisher (para$ara), do thou vex (ard) away the vehemence 
(qusma) of them ; then get us wealth. " I think d rather means, 
e do thou make their vehemence fall far away (4. ii) '. 

(i) In the defeating or killing of enemies the idea of making 
them fall (4. iii) is prominent. 

( j) 6irsna Siro 'psasa 'pso arddyan amsun babhasti hdritebhir 
asdbhih, 6. 49. 2. Whitney has ' exciting (ard) head with head, 
breast (dpsas) with breast ', etc. It rather means ' pressing (5) 
head with head ', etc. Ardati in this sense is common in later 
Sanskrit. 

Rdantu in the verse under discussion appears to contain the 
root-aorist stem and to mean 'fall' (2. ii). Accordingly our 
verse may be translated : " Indra, slay the male sorcerer and 
the female, shining forth with magic ; let the root -devotees fall 
neckless (on the ground) ; may they not see the rising sun." l 

em- 

udapruto Marutas ta iyarta vrstir ya viSva nivdtas prnAti : 
ejati gldhd kanye 5 va tunndi ' rum tundand pdtye ' va jay a. 

6. 22. 3.=TS. 3. 1. II 8 , with variants : Marutas 

vfstim y viSve Maruto jundnti : kroSati gar da p6rum 

tunjana 

V. 1. : Ppp., with the majority of SPP.'s authorities and some of Wh.'s 
(P.M.), reads udaplutas. One of Wh.'s (W.) msa., three of SPP/s, and 
apparently also the commentator's text have gdlha for gldha. See Wh.'s 
notes. 



1 The meaning * to shine forth, to become prominent ', given by 
GrasBmann. as the first meaning of v /Md-, seems to suit all the Rigvedic 
occurrences. Muardevah are those who are devoted to the roots, as means 
of sorcery, as the ^sis to the gods. It does not necessarily imply literally 
worshipping the roots, but simply expecting everything through their help, 
as others do through that of the gods. For Fuller discussion, see . 
below. 



VOL. XVII. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 55 

Pada ttin ytih vWt>5& : p&tya iva. . . . 

Previous Scholars : SSyana on AV : " he Marutah udapnUah udatoya 

prerakan tan meghan iyarta prerayata ya yadiya yesam meghanarh 

savhbandhini vfatih vtiva vitivani vrlhiyavadini nivatah nimnagaminir 

nadtt ca pfnati purayati: gahlayati kwtsayati bHltim utpadayatlti 

gahla stanayitnurupa madhyamika vak : yathS tunria daridryadi- 

bhih pldita Icanya matapitradln kampayati tadvat erum gantaraih 

megham prapya tuftjana abhasamana dhvananti. . . .patya sdhita jayeva. ..." 
The commentary on the TS. is essentially the same except for slight 
modifications required by the text. Pischel, Ved. Stud., I. 81 -85, has a long 
discussion on this verse, and after trying to show, tatdm; gaTda- 
synonymous with gldha-, both adj. meaning 'lascivious'; p6ru-<^/pi 
and m-< v /*r, both synonymous, meaning * penis', and tuftjand or 
tundaria, passive he translates : " O ihr Maruts, im Wasser schwimmend, 
sondet solchen Regen herab, dass er alle Thaler Anfiille. Er moge he cab 
stiirzen wie ein geiles Madchen (sich heftig bewegt), wenn sie gebraucht 
wird (tunnti), wie eine Frau, wenn ihr von dem Mamie der penis eingestos- 
sen wird." Whitney renders, "Water-swimming are the Maruts; send 
ye that rain which shall fill all the hollows ; the gldha shall bestir itself, 
like a girl that is thrust, thrusting the 6ru, like wife with husband," and 
remarks, " The text of this verse is hopelessly corrupt, and all attempts 
to make connected sense of the second half must apparently be ( like that 
of Pischel in Ved Stud. I. 81. if.) forced and unsuccessful". See his 
elaborate notes. 

It seems necessary first of all to try to reconstruct the ori- 
ginal text and to find out the value, grammatical or otherwise, 
of the obscure words. Udapruto Marutas of TS. may be 
accepted, the first qualifying nivdtas and the second being a 
vocative. Ta stands for tah, qualifying nivdtas, the Anunasika 
standing merely to prevent a hiatus, cf . aminantft evaih RV. 1 . 
79. 2, Icsa nd bhisd adrivah ib. 1. 133. 6, siriyah satis tti (p. ta) 
u me pumsd ahuh ib. 1. 164. 16, etc., and Wack. I. 267. y. 
The b of TS. is simply a substitution, due to reminiscence, of 
b of RV. 5. 58. 3. Gldha is a simplification to a better known 
word of gdlha (see v.l. above), which must have come out 
of gdlda, a variant of gdrda, through a confusion of d (<) 
and h ( ^r ) of . hasydn dhasyan in some mss. for dasydn ddasyan 
AV. 6. 71. 3 (in Wh.'s collation book), ahuta, variant of ad uta 
AV. 19. 2. 5 (Wh.'s notes), etc. firu- in AV. must be a corrup- 
tion for p6ru, occasioned through a confusion of e ( *r ) and p 
(* ), cf. v.l. to Parisistas of the AV., XLVIII. 116, where mss. 
A.B.T.V. read pelava for ailava ;_ Oh. 8. *g i wnr for ^g I 



56 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

AV. 8. 8. 3, and footnote to dlpaayu. Kr66ati in TS. is due 
to a remembrance of Nigh. 1. 11, where galda- is considered a 
synonym of vac-. Pdtyeva is for pdtya iva < pdtye iva, an 
instance of double Sandhi, cf. vdndaneva for vdndanah iva AV. 
7. 115. 2, krtyeti for krtye iti 10. 1. 15, vateva for vatah iva, 
ucchisaisdm for ucchisah esam (as shown by meter) AV. 10. 1. 
17, etc. Jayeva pdtye is a favourite Vedic expression, here 
inverted for the sake of meter, cf . jayeva pdtye tanvdm riricydm 
RV. 10. 10. 7, which contains the same idea as here, and also 
RV. 1. 124. 7, 10. 71. 4. Or, probably va is to be read here 
instead of iva as in a number of instances in the RV., cf. Grass - 
mann, Wort., s.v. iva. Tunjana as read by the commentator 
and the TS. will be found a better reading than tunddna. 

Thus the reconstructed text would be (in pada-patha) : 
udaprutah Marutah tah iyarta vfstih yah vi&vdh nivdtah 
prnali : ejdti gdlda kanya iva tunna perum tunjana pdtye iva 
jdya. 

We have now to find out the meanings of gdlda- and pern-. 
Kesavasvamin, a very careful and well-read lexicographer (see 
Introduction to Kalpadrukosa, GOS. XLII, vol. T, p. 
xxxviiif .), gives the meaning * stream or current ' to garda : 
garda sin dravadharaydm dhamariibu ca vdci na. The other 
two meanings, viz.. ' veins ' and ' speech ', must have come out 
of a metaphorical use of the word, ' veins ' and ' speech ' con- 
sidered as streams. This is apparently supported by the 
Nighantu, which, besides counting galda- among the homony- 

mous words at 4. 3, has at 1. 11 : &lokah \ dhard \ 

dhamdm \ ndlih \ galda \ sdrah \ suparm \ bekure 'ti 

saptapanca6ad vdnndmdni. Even a cursory glance at the 
Nighantu will suffice to show that the words given there as 
* names ' of a certain thing are not to be taken literally as 
synonyms. They are in almost every case a motley collection 
of synonyms as well as of words adjectively or metaphorically 
used in reference to the object in question. The same must 
be the case here, as the selected words given above will show. 
Nir. 1. 6. 24 has, (( galda dhamanayo bhavanti galanam asu 
dhlyate : ' a tvd vifantv indava a galda dhamdnmdm ' ( found in 
M. Sr., and elsewhere with variants : see Ved. Concord.) ; ndnd- 



VOL. XVII. FT. I.] INTERPRETATION OF DOUBTFUL WORDS. 57 

vibhakty ete bhavatah, agalana dhamaninam ity atra 9 rthah. " As 
a matter of fact, the meaning c stream ' is what is required here, 
c let the drops (of soma) enter thee, enter the streams of thy 
veins'. The other known occurrence of the word is in RV. 
8. 1. 20=SV. 1.4.1.2.5: ma (a, SV.)tva somasya gdldaya 
sdda yacann ahdm gira (jya, SV.) : bhurnim mrgdm nd sdvanesu 
cukrudham kd Uanam nd ydcisat. " May I not, asking always 
(for something) with streams of soma, (and) with prayer, anger 
thee in the libations, like a (wild) boast ; who would not beg of 
the lord?" Cf. somasya dhara RV. 9. 80. 1, and many other 
instances where dhara- stands for the stream of soma, Grass- 
mann, Wort., s.v. dhara. The above two will, I think, explain 
why the Nigh, grouped dhamdni- and gdlda- among the names 
of speech. H. gad f . ' impure liquor from an indigo vat ', Beng. 
gad ' foamy dirt on the surface of a liquid, such as treacles', are 
probably the same word as gdlda-, the meaning having con- 
ceivably been transferred from the foamy surface of a stream, 
when full. 

Peru- appears to be only a bye-form of pera- or pela- ' testi- 
cles '. For alternaces of a and u 9 see above under dksu. As a mat- 
ter of fact Vj. 178, 126 has pheluka-, astriyo muskako6andak 
2>heluko vrsano 'ndukah. Here it stands for the female genera- 
ting organ ; cf . muskd ' testicles ' used in the same sense, 
amusya ddhi muskdyoh AV. 6. 138. 5, arayan asya muskabhyam 
bhdmsaso ' pa hanmasi 8. 6. 5. Transference of meaning from 
{ testicles ' to the * generating organ ' is also found in Hindi Sr, 
Beng. er 'penis' < anda- 'testicles'. For use of singular 
instead of dual, cf. urum in AV. 14. 2. 39, a roho 'rum upa 
dhatsva hdstam, etc. Kautsavya Nighantu, Parisistas of the 
AV., p. 315, 124, has f paramgativilike (v.l., tilike, puramgati- 
vilike, paramgatl )] ill striprajananasya, which, as it stands, 
does not give any sense. As this Nigh, professes to be particu- 
larly attached to the AV., we may expect to find these words in 
the latter. The correct reading may probably be perv-anga- 
tauvilika iti, etc., of which pent- is found here, referring to the 
female organ (striprajanana-), dngena in eva te $cpah sdhasa 
'yam arko 'ngena 9 ngam sdmsamakam krnota 6. 72. 1, and 
tauvilika in tauvilike 9 ve 9 laya 9 va 9 ydm ailabd ailaylt 6. 16. 3, 
whose meaning is not certain, but may possibly be the same. 



58 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

Thus we may translate : " O Maruts, impel those hollows 
(viz., rivers, lakes, etc.), over-flowing with water, all of which 
the rain fills up : may the current heave like a maiden, struck, 
(or) a wife offering (lit., thrusting forth) her member for the 
husband." 

The comparison is between the undulation of water, 
stirred by the wind, and the rising and falling of the breast 
of a girl, sobbing when she is beaten, or of a wife, due to the 
agitation in coitus. For tunna, cf. gravna tunno abhistutah 
pavitram soma gacchasi RV. 9. 67. 17, and for tunjana, cf. dthA 
'bhara Syena-bhrta prdyamsi rayim tunjano abhi vajam arsa RV. 

9. 87. 6. For a similar idea as ind, cf. RV. 5. 61. 3, 10. 10. 8, 

10. 86. 6, etc. 

kdkulsala- 

a-sau ha ihd te mdnah kdkutsalam iva jamdyah : abhy enam 
bhuma urnuhi. 18. 4. 66. 

Previous Scholars : Tho commentator roads ktikutsthala- , explaining 
it either as the head or the part of body just below the neck. " Dieser 
[N".] du hast deinem geist gelaszen [kakutsalam ? kutsalam es 1st urn eine 
silbe zuvil ; jamayah kanu nicht richtig soin, wenn es von jami stammeii 
soil. Villeicht ist es ja-inayah] als erdegebildeter den im haupte wandehi- 
den [kakut-sala] | , bedeck ihn rings, o erd||". Luclwig, p. 492. "Thou 
yonder, ho ! hither thy mind ; as sisters (jami) a kdkutsala, do thou cover 
him, O earth". Whitney. "The translation implies the evidently 
necessary emendation to dsau in a ; both editions give asau because this is 

read by all mss. The comm. understands the word as vocative The 

Pet. Lexx, conjecture kdkutsala to be a pet word for a little child". ib., 
notes. 

The correct reading seems to be kdkutstfwla as read by the 
commentator, the tti having been dropped by manuscript 
corruption. Such dropping of the last of three consonants 
coming together is found elsewhere, cf., for instance, dr6d- for 
ar&yd- at 4. 4. 5, and krdmasvdrSa for -r$ya at v. 5, ksinkah for 
ksvi- in two of Wh.'s mss. at 8. 3. 7, dhuksa for -ksva in several 
mss. at 10. 9. 13d M budhnat for -dhnyat in three of Wh.'s at 4. 1. 
5, etc. Kakut' t usually * the hump of a bull ', also signifies any 
raised place or prominence, and in kdkutsthala- it may stand for 
* the buttocks of a woman', as it certainly does in kakudmati- 
' waist', lit., 'having a hump*. The difference in accent is 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 59 

undoubtedly due to the unintelligibility of the word. This 
meaning would give a point to the comparison, suggesting a 
complete and careful covering of the bone-relics to which they 

refer : " like women their buttocks, do thou cover him 

round. earth". 

kandknaka- 

ydd agnau surye, visdm prthivyam osadhlsu ydt : kandavisdm 
kandknam niraitv aitu te visdm. 10. 4. 22. 

Previous Scholars : The commentary is wanting. BR. consider each 
of kandavisd- and kandknaka- a different variety of poison. Ludwig, Henry 
and Whitney do not translate the two words. Bloomfield, who agrees 
with them in not translating, remarks afc p. 608, " Kandavisham and 
kanaknakam are CITT. Aey. ; it is not even certain that the latter refers to 
a particular substance : the word may be an adjective qualifying kanda- 
visham. It seems to be an intensive formation from root Kan. " 

As suggested by Bloomfield, kandknaka- appears to contain 
an intensive stem of root kan ' to be bright '. The last -ka is, of 
course, a diminutive suffix. The a- vowel, instead of i, between 
the two parts of the reduplicated root is as in caracard-, 
calacald-, ghanaghand- , etc., while its lengthening is prevented 
by the following conjunct, just as in the case of the i- vowel, cf. 
Whitney, Sans. Gram. 1002. IHf ; Brugmami, Grund. 467. The 
syncope in the second part is just like that in pdnipnat- : pan-. 
Thus kandknaka- would mean ' bright ', * glistening ', while 
kandavisd- evidently means * root -poison ' (\kanda- 'root', for 
length of the second vowel cf. sahdsramagha-, d$vamagha-, etc.). 
I should think from the first hemistich that the former refers to 
the poison (of a serpent) that is in the fire or in the sun, which 
would, of course, be bright, while the latter to that on the 
earth, among the herbs. 

kamdla- 

ydlt krnoti mrtdvatsam dvatokam imam striyam : tdm osadhe 
tvdm naSaya 'syah kamdlam anjivdm. 8. 6. 9. 

Previous Scholars : The last pada has been understood variously : 
'ctsyah kamdlam garbhadvaram anjivam abhivyaktimat mlaksanopetam 
v5. s Say ana; ' und ihre scheide sei glatt,' Ludwig, p. 624; Tdtre 
lubrique et glissant qui la convoite,' Henry ; "Whoever makes this 
woman having a dead child (-vetted) or a miscarriage, him, O herb, do thou 
make disappear, lustful [accusative] for her, slippery." Whitney. 



60 lOTERPBETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

Kamdla here appears to be the same word as 6amara- 1 
in GBr. I. 2. 18, tasya ha snatasya '6vasya 'bhyuksitasya 
roma&amarebhyo (cf. romakupa-) 'ngara aSiryanta, and I. 5. 5, 
etavanta eva purusasya pe6a&amarah, where it evidently means 
'hole', 'pit', or 'cavity'. It is apparently also identical 
with Gr. Kapdpv) ' vault ', 'ear-hole', etc., Lat. camera 'bent ? , 
old Pers. kamara ' girdle ' Meyer, Handbuch d. Griech. 
Etymologie, Goth, himins, old H. Germ, old Sax. himil, 
'heaven', originally 'vault' Boisacq, Greek Dictionary, 
s.v. Kapsdprj. 

As words denoting ' cavity ' or ' hole ' such as bhedd- (RV.), 
kuhara-, vivara- (later Sans.) are often used to denote also 
the 'pudenda' or 'womb', the commentator is apparently 
right in his conjecture, as the context also shows. Anji- in 
anjivdm and in VS. 23. 21, titsakthya dva guddm dhehi sdm 
anjim caraya vrsan : yd stnncim jivabhojanah, seems to stand 
for ' the seminal fluid ', not ' penis ' as given in the dictionaries, 

kalmali- 

Sraddha pumScali Mitro magadho vijnanam vaso ' har usni- 
sam rtttrl ka hdritau pravartau kalmalir manih. 15.2.1(5). 



1 Altcrnaiices of v and k, of which the definite conditions aro un- 
known, are found in a number of case^. Wack. 1. 201. a. gives the 
following instances : nisant- * bright' : rue- ' to light ' ; sru- * hear' : ktirna- 
* ear' ; &ram- ' to be tired* : klam-, klanta- ; lopaAA- ' jackal' : lopaka- ; *!- 
'to crush,' tifnatl, Slrna- : JDhp. krnati, klnia- * to harm,' * to kill". 
To these may be added: krakasa : krakaca- * saw * ; yuvajd'.' yuvaka- 
' young man,' and, in general, suff. -6a (ttaxa-, babhluta- etc.) : suff. -ka : 
sabala-: kabara-, Magha 5. 19, Halayudha 4. 56, 'variegated'; fambara-: 
kambala- l a kind of deer ' ; &ambu-, ,<ambnka * snail ', ' rice dust ' : kambit-. 
kambuka-, ' conch,' 'rice dust' (Vj.) ; element 3ar- in sarvarl- * night", 
^ra-, sa(6a1)raiiga- ' variegated ' , element kar< in karvara-, karvura-* 
kalmam-, * variegated ', karvarl * night/ Ujjvala, Unadi. 2. 123, and sir- in 
4ilp&- ' variegated ', dirina- * night ' : kir- in kirmira- * variegated,' (all : 
V6rai~, I.-E. kera* (Waldo) 'to mix,' 'to cook'?); sarabJia-: karabha 
' young elephant ', 'camel'; &arkam-: karkara- 'gravel'; 6ala-: kala- 
' resin,' Kalpadruko^a, G6, 415; tirana-: kana- 'one-eyed', ib. 51, 279; 
kantha- ' quilt ' : tnath- ' to pierce' (?) ; kdrudatin- * having broken teeth' : 
*6arus-dat, root rff- 'to crush ' (?) ; tampa- 'lightning': kamp- 'to 
quake ' ?, cf . capala-. As variae lectiones, koka- for kosa- ' name of a 
river ', viklrna- for vidirna-, BR. s.v. sar+vi, etc. 



VOL. XVn. PT. I.] INTERPRETATION OF DOUBTFXTL WORDS. 61 

usali pum6caK mdntro magadho vijnanam, etc., v. 2(14). 
ira pum6cali hdso magadho vijnanam, etc., v. 3(19). vidyut 
pum$cali stanayitnur magadho vijnanam, etc., v. 4 (25). 

pro, babhrdve vrsabhdya 6vitice maho mahim sustutim iraydmi : 
namasya kalmaUkinam ndmobhir grnlmdsi tvesdm Rudrdsya 
nama. RV. 2. 33. 8. 

tvdm Indra Sarmdrina fiavydm p&ravatebhyah : vipraya 
stuvate vasuvdnim DuraSravase vaha. AV. 20. 135. 11. (Sarmd 
rinah, RV., GB., AS., SS., AB., KB., of which only 66. has 
the whole verse, others only the first pada. Ved. Concord.). 

Previous Scholars : Sayana on RV. : ' jvafato namadheyam etat (Naigh. 
1. 17) : jvalantam : kalayaty apa gamayall malam iti kalmallkam tejah : 
tadvantam.' Grassmaim, * otwa bunter, funkelnder Glanz.' Whitney. 
"....faith is the harlot, Mitra the magadhd (bard ?), discernment the 
garment, day the turban, night the hair, yellow the two pravartds, kalmali 
the jewel (matu) ", etc. 

In form Sarmdri- in AV. 20. 135. 11 appears to be identical 
with kalmali-, see discussion under Icamdla-. Now, Sarmdri- is 
used as an instrument of Indra and kalmali" in the RV. verse 
as something possessed by Rudra. Both these gods have the 
weapon vdjra 'thunderbolt' in common. The first group oi 
verses from the AV. relate to the paraphernalia of Vratya as 
those of a king. As all the rest of these refer figuratively 
either to abstract qualities or to natural objects or phenomena, 
hdritau and kalmali- should be expected to do so too. Harit- 
f .du. in RV. 3. 44. 3, dyam fndro hdridhayasam prthivim hdri- 
varpasam : ddhdrayad harilor bhuri bhojanam ydyor antdr hdri& 
carat, probably refers to the two worlds (so also Grassmann). 
The same may be the case here, as pravartau (found also in 
Ap. r. 19. 23. 11, 13. 24. 2) means < ear-rings' (Sayana on 
TS. 2. 3. II 4 , quoting Ap. r.). ' Thunderbolt ' for kalmali- 
(or sarmari-) would suit here as in the other two passages and 

Whitney's translation may be modified, " the two worlds 

the ear-rings, the thunderbolt the jewel." AV. 20. 135. 11, 
with the obviously required emendation of vaha to -has and 
DuraSravast to Dura-, may be translated: "Thou, O Indra, 
with the thunderbolt, broughtest from the Paravatas the 
desired (havyd-) acquisition of wealth for the wise, singing, 
Durasravas. " 



62 INTERPRETATION OF DOUBTFUL WORDS. [.T.B.O.B.8. 

Kalmali- is probably connected with the group of words 
containing the element, 6ar-, Sir-, kar-, kir- (such as Sarvari-, 
kalmasa- etc., see footnote under kamdla-), all meaning ' varie- 
gated ' and its original meaning might well have been ' many- 
coloured ', * shining ' (cf. the epithets, darsatd-, dyumdnt-, Subhrd-, 
hdri-, hdrita-, hiranydya-, of vdjra, Grassmann, s.v. vdjra). 

kumba- kurira- opaM- 

tvdm virudhdm SrestJiatama 'bhi&ruta 'sy osadhe : imam me 
adyd purusam klibdm opaMnam krdhi. 6. 138. 1. klibdm krdhy 
opa6inam diho kunrinam krdhi : dtha 'aye, 'ndro gravabhyam 
ubM bhinattv andyau. v.2. kliba kUbdm tvd 'karam vddhre 
vddhrim tvd 'karam drasd 'ra&dm tva 'karam : kuriram axya 
Airsdni kumbam cd ' dhinidadhmasi. v.3. 

stoma asan pratidhdyah kuriram chdnda opa&dh : Suryayd 
ASvina vara 'gnir asit purogavdh. 14. 1. 8.=R/V. 10. 85. 8. 

Simvali sukaparda sukurira svaupa$a. TS. 4. 1. 5 8 .=VS. 
11.66. 

Kumba- and kurira- are also found in AP. Sr. 10. 9. 5, 6, 7 
(and in the corresponding siitras in Baudh. Sr. 6.1 : 6. 4, 5 ; 
15. 15 ; 25. 4) : atha patmfrirasi kumbakunram adhyuhate. 
krsnam jwornan'dm iti Vajasaneyakam. jdlam kumbakuriram ity 
dcaksate. Kumba- alone is found also in AP. Sr. 1. 21. 3, udid- 
nakumbdm frimyam, where it has a development of the original 
meaning. Kurira- is found also at AV. 5. 31. 2, yam tc cakruh 
krkavakdv aje va yam kunrini : dvyam te krtyam yam cakruh 
punah prdti liar ami tarn, where it has a different but not 
wholly unconnected meaning. Opa6d- is found also in AV. 9. 
3. 8, discussed under dksu: and in RV. 1. 173. 6, prd ydd 
ittha mahina nfbhyo dsty dram rodasl kaksye na 'smai: sdm 
vivya tndro vrjdnam nd bhuma bhdrti svadhavfi opa6dm iva dyam ; 
ib. 8. 14. 5, yajnd tndram avardhayad ydd bhumim vy dvartayat : 
cakrand opoSdm divi\ ib. 9. 71. 1, a ddksind srjyate Susmy 
dsadam veti druho raksdsah pati jagrvih : hdrir opa&dm krnute 
ndbhas pdya upastire camvor brdhma nirnije; Tand. Br. 4. 1. 1, 
gavo va etat satram asata tasam da$asu massu sfngdny ajdyanta 
ta avadann arasmo 'ttisthamo 'pa&a no 'jnate ' ti td udatisthata, 
and 13. 4. 3, dvyopa&ah samstuta bhavanti tasmad dvyopa$ah 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 63 

pafavah; and in go-opa&a-, epithet of dstra-, RV. 6. 53. 9 and 
hrdayaupad- VS. 25. 8, in all of which it has a slight modifica- 
tion of the original meaning. 

Previous Scholars : Sayana on AV : ' upaete asmin purusa iti opa6ah 
strwyafijanam', 6. 138. 1; *kurlrah ketah', v.2; 'fewfraw keSajalam 
kumbam tadabharanarii ca strmam asadharanam', v.3. In explaining the 
passages in RV. he seeks the help of etymology and arrives at a different 
meaning of opa6d- in each case, thus: sffigam or lokadvayaml. 173. 6.; 
upetya Sayanam or vlryavUesah 8. 14. 5 ; sarvasya dharakam 9, 71. 1 ; 
upaSerat ity opatiah, gava opasa yasyas tad%s* 9 ata eva paAu-sadhanl, 6. 53. 
9 ; and goes farther at 10. 85. 8, by stating that there is a metre called 
kurira-, 'suryaya rathasya stomas trivfdadayah pratidhaya asan : pratidhiyata 
iti pratidhaya watiryagayatakasthadayah : latha kurlram chandafy kurlra- 
namakam chando 'nasa opao 'bhavat: yeno ' patierate sa opaah\ BR : 
kumba * eine Art weiblicher Kopfputz ', kurira= ' eine Art Kopfschmuck 
der weiber', opad- = 'ein Kopfutz : Biischel, Locke, cirrus; oder 
viell. Zoff 5 and 'Flechte' at AV. 9. 3. 8. Geldner, Ved. Stud. I, 130137. 
basing his arguments on AV. 5. 31. 2, Tand. Br. 4. 1. 1, 13. 4. 3, and Ap. 
r. 1. 21. 3 (where he wrongly interprets kumba- as the 'pointed end'), 
concludes that all the three words mean in the first place ' horn J and then 
a ' horn -shaped head-ornament ' or * diadem '. Oaland alone, having had 
access to Baudh. Sr. and the commentaries, has given the right interpreta- 
tion to kumba- and kurira- in his translation of Ap. Sr. 10. 9. 5. 

Baudh. Sr. 25. 4 (p. 232, 1.1.) has, vidalam u ha kumbam 
bhavati jalam u kuriram, and thereupon the commentator adds 
vam6avidalam jalasya nemibhutam as kumba-. The commentator 
to Ap. r. 10. 9. 7 also has, jalam anayah, lac ca vaidale valaye 
syutam, and quotes Baudh. for his support. Thus kumba- is a 
circular rim made of sliced bamboo and kurira-, a net of sheep's 
wool (Ap. &r. 10. 9. 6, above), of which the kumba- formed the 
edge. The two together formed a dome-shaped female head- 
dress (that is, when put on) and are often referred to as one, cf . 
kumbakunram Ap. Sr. 10. 9. 5, 7 above and Baudh. &r. 6. 5, 
15. 15. 

Kumba- ( sacrificial enclosure ' must be the same word as 
kumba- with an extended meaning. It seems to contain a root 
kum- ' to bend ', probably to be traced in komya- RV. 1. 171. 3, 
stutaso no Maruto mrlayantu y td stuto maghdva 6dmbhavi8thah : 
urdhva nah santu komya vdnany dhani vi&va Maruto jigwa, 
which Sayana explains as kamamyani 'pretty' and others 
have hesitatingly followed without being able to find a satisfac- 



64 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

tory derivation. If, however, komya- contains this root, it 
would mean 'bending', 'pliant', which as applied to trees, 
desired to be saved from the wrath of the storm-gods, would be 
most appropriate (the third pada giving the sense, ' let our 
pliant trees remain erect'). The root is probably also to be 
seen in kumbha- ' pitcher ', Gr. icu/^/fy, Kvpflos (for formation 
cf. kamba-, kambha-: kam-, Pan. 5. 2. 138); in kumara-, 'the 
bending, crawling, baby', in komala- ' bending ' > 'yielding' > 
' soft', etc. ; and in the root kunc- ' to bend ', which may be an 
enlargement of kum-. In view of Lith. kumpti ( to bend one- 
self, kumpas 'bent', Lett, kiimpt 'to be bent', old Pruss. 
etkumps adv. ' round about ' ( Walde, s.v. qam-, qamp- e to bend '), 
Gr. Ku/xj8os 'pitcher', Kvpfiaxos 'upper part of a helmet' 
and of Gr. Ka/xrrra> ' to bend ', Sans, kamdla-, and its correspond- 
ences in the other languages given under it, there seems to have 
been a duplicate root *kam or *kum- in the original language, 
which gave rise to the various enlargements and derivatives. 

From looking upon kumba-kurira- as one thing kumba-, 
and probably also kurira-, came to denote the whole thing, which 
as we have already remarked looked like a dome or vault when 
put on. Thence came the use of kumba- as the 'knob' of a 
stick or peg of a yoke in Ap. Sr. 1. 21. 3, udicinakumbam Samyam, 
the commentator explaining kumba- here as the thick end of a 
Samya-. As already in the time of Baudhayana and Apastamba 
the use of this form of female head-dress had become obsolete 
(judging from their manner of describing it, of course,) it 
may well have been Indo-European and Gr. Kvpfiaxos ' the 
upper part of a helmet ' may have the same source. 

Kurira-, which as we have seen was made of sheep's wool, 
meant also 'blanket', cf. kuriras tu puman malaviSese kambale 
' pi ca : klibam tu maithune pad/me, jale ca munibhasitam, 
Kesavasvamin's Nanartharnavasamksepa, Triv. Sans. Ser. It 
might well have originally meant 'wool' and in AV. 5. 31. 2, 
quoted above, kuririn- wedged, as it is, in between aji ' goat ' 
(loc.) and dvyam ' ewe ' (loc.), may possibly stand for ' sheep '. As 
a matter of fact we find lewrari- (for kuriri ?) and jalakirii- (lit., 
'furnished with a net') as names of an ewe, see VaijayantI 
70, 129. But the case is doubtful, and kuririni may be an 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 65 

adjective to aj6, kurira- referring to its horns as 'head-orna- 
ments '. This last remark applies with greater force and more 
certainty to opa&d- in Tand. Br. 4. 1. 1 and 13. 4. 3, quoted 
above. 

A comparison of AV. 6. 138. 1, 2, 3, RV. 10. 85. 8 = AV. 
14. 1. 8, and TS. 4. 1. 5 3 = VS. 11. 56 will show that either 
kurira- and kumba- or kurira- and opatd- are found together, 
but nowhere kumba- and opa6d-. This, as referring to a parti- 
cular female head-dress, is sufficient to give us kumba- = opa&d-, 
' the circular sliced-bamboo rim'. This, as we have seen under 
dksu-, is its meaning also at AV. 9. 3. 8. In AV. 14. 1. 8= 
RV. 10. 85. 8, pratidhi- probably means * ornament' as * some- 
thing put on' (compare also Wh.'s remark under AV. 14. 1. 8) 
cf . prdti ydd asya vdjram bahvor dhuh RV. 2. 20. 8c 3 and the 
verne may be rendered, " The ornaments were the laudations, 
the net and the rim, the metre ", etc. OpaSd- in RV. 1. 173. 6. 
8. 14. 5 and 9. 71. 1 refers to the whole head-dress, and with 
this modification the translations of Geldner in Ved. Stud., 
1.131f ., may be accepted. G6-opa6a- probably means ' with 
the knob made of cow's bone (or covered with cow's hide) ' and 
hrdayaupa&d- l that which covers the heart '. 

kurutini- 

aydm pdnthah krtyeti tva nayamo 'bhiprdhitam prdti tva prd 
hinmah ; tend 'bhi yahi bhanjaty dnaavati 'va vahim viAvdrupa 
kurutim. 10. 1. 15. 

Previous Scholars : The commentary is wanting. BR. think that it is 

probably = kirltini-. Ludwig : " auf diesem wege geh brechend los 

zum angriff, wie ein vollstandig heer mit wagen und mit rossen [ ?]." Bloom- 
field : ** go this way like a crushing army, with heavy carts, thou 

that art multiform, and crowned with a crest ^?), " and at p. 604, 
'* Kuru^ini translated by * crowned with a crest ' is in truth a an. Aey. of 

unknown value HemaA;andra also reports a word kurudn ' horse % 

and Ludwig apparently, on this basis, translates *mit rossen'". Henry: 
'par ce chemiii marche en te le frayant, comme monte sur un chariot, 
mont^e sur un char, rev&tue de toutes les formes, coiffde d'une mitre '% 
and in the commentary, "vahinl synonyme de dnasoati, et, quant a 
kurutim, cf. Tusuel kiritin et 1'epithete tiritinas AV. VIII. 6. 7." Whit- 
ney : *' breaking, tike a draft -cow with a cart, all- formed, wearing 

a wreath (? kurutln) ". 



66 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

Kuruta- in kurutinl seems to be identical with later 
Sanskrit kurula- or -rate- 1 'curls on the forehead (bhramarala- 
ka-)\ found also in Prak. (for correspondence of Sans. t>d>l 
in middle Indian see references under arataki-). Kurutinl would 
thus mean ' having curls on the forehead ', and this would 
fit in with krtya- ' witchcraft ', which has been described several 
times in the same hymn as a woman 2 with head, ears, nose, 
etc., cf. v. 1, yam kalpdyanti vakatau vadhum iva viSvdrupam 
hdstakrtam cikitsdvak : sa 'rad etv dpa nuMma endm ; v. 2, 
Sirsanvdti nasvdti karnini krtyakfta sdmbhrtd vUvdrupa, etc. 

For kurutin- 'horse' in Bohtlingk's edition of Hema- 
candra's esanamamala, v. 176, the Bhavanagar edition of 
AbhidhSnacintamani gives kutara-. If the former reading is 
correct, it may signify a ' horse with a tuft on the forehead.' 

Padas c and d, which have been diversely translated, 
seem to compare the retreat of krtya- to that of an invading 
army (cf. bhagna-, 'vanquished': bhanjati). Anasvati, 'fur- 
nished with chariots', epithet of vahint 'army', is probably 
due to a reminiscence of vahatau vadhum iva in v. 1. 

khadura- 

khadure 'dhicankramam khdrvi/cdm kJiarvavasintm : yd 
udara antdrhita gandharvapsardsa^ ca yt : sarpa itarajana 
rdksdmsi. 11. 9 (11). 16. 

Previous Scholars : Sayana : ' durabhutarh kham klwtduram akae 
duradeeS Ludwig. p. 531 : " die tiber den khadiira [das sohwert] taumelt 
die verstammelte, bei verstQmmelten [auf dem schlaohtfeld] wonef, 
etc. Henry : Celle qui monte sur le brancard mortuaire ", etc., and 
in the commentary, " II semble que khadura doive necessairement s'inter- 
preter par le posterieur khadu." Bloomfield, who renders with *mist', 
remarks at p. 636, ' Our rendering of khaehire reflects simply our own and 
Sayana's perplexity.' 

Khadura- appears to be identical with later Sanskrit 
khalura- in khalurika- c place of exercise for soldiers '. This 
meaning suits the context here : the whole hymn has reference 
to battle, soldiers, and weapons, and a spectre haunting the 

1 The long u in kuru^im may be due to the exigencies of metre. 

2 Cf. Kurulalikulavalikhyamanabhrulatantahrdayahgamarh, epithet of 
a priyatamajana, Somadeva's YaSastilaka I, 526, 2. 



OL. XVH. PT. I.] INTHEPBBTATION OF DOUBTFUL WORDS. 67 

place of exercise may be expected to be mentioned amidst a 
host of other different kinds of spectres. D is regularly re- 
presented by 1 in RV., and for d>l (1), in Pali and Prak., see 
references given under arataki. 

khfgala- 

pMnge sutre khfgalam tad a badhnanti vedhdsah : srava^yum 
Auamam kabavdm vddhrim krnvantu vandhurah. 3. 9. 3. 

navk 'va nah pdrayatam yug& 'va ndbhye 9 va na upadhiva 
pradhiva: 6vane 'va no drisanya tanundm khfgale'va visrasah 
pdtam asman. RV. 2. 39. 4. 

Previous Scholars: Say an a on both the passages explains khggala- 
as 'armour'. BR., 'viell. Stab, Kruoke'. Bloomfield, 'talisman', but 
agrees with BR. in the meaning * orutches ' for RV. Whitney does not 
translate the word, 

Kausika, 43. 1, has concerning this hymn (AV. 3. 9.), kar6a- 
phasyeti piSangasutram araludandam yadayudham, which, when 
compared with the verse quoted above, gives us khrgala- = 
danda- f stick'. This is supported by the RV. verse quoted 
above, its fourth pada meaning, * like two sticks save us from 
fall'. Khrgalya- in MS. 2. 7. 12 (=Ap. Sr. 16. 18. 4), udyojan- 
nam antaryamam tsam khrgalyam (Ap. kha- or khadga-) avam 
(Ap. 6apham) : astrdm tddam (Ap. -la-) pratmdhd ubhe mandu- 
kyau yuje, apparently means the same thing. 

galuntdh 

asutikd rdmdyany ctpacit prd patisyati: glaur itdh prd 
patisyati sd galunto natiayati. 6. 83. 3. 

Previous Scholars: Sayana: ' glauh vranajanito haraakaayali itah 
asmad angat prapatisyati....yadvci glaus candramah....pragamayi8yati.,.. 
aa candramah. galuntah. gandamxlodbhavavikarena tatra tatra hastapadadi - 
eandhisu udbhutan gadun tasyati upaksayatlti gaduntah ', etc. Ludwig : 
" ....fort von hier die eule [glauh] fliegen, und der vogel wird versohwin- 
den." Bloomfield, p. 17 : " ....the boil will fly away from here, the galunta 
(swelling) will perish". "Barren shall the apaclt, daughter of the black 
one, fly forth ; the boil (glau) shall fly forth from here ; it shall disappear 
from the neck (? galuntds)". Whitney, who remarks, "the translation 
here given of galuntds is the purest conjecture, as if the word were a 
corruption of some form of gala (our W.O.D. read galantds), with ablative 
suffix tas ". 



68 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

If galantds, the reading of three of Whitney's mss., be 
correct, its relation to root gal- ' to ooze ' would be just like 
that of jivantA- living one', AV. : jiv- 'to live', tarantd- 
n. pr. : tf ' to conquer ', pantd- ' drink ' : pa- ' to drink ', vasantd- 
' spring': vas~ 'to be bright 5 , vesantd- 'pool 5 : vis- 'to enter 
(e.g., running water) ' (?), hemantd- ' winter ' : him, himd- ' cold ', 
' frost ', etc. Being thus presumably a substantive like the 
others, it would mean in this connection ' the oozing one \ 
viz., 'the pus'. If Sayana's explanation of aautika- as 'not 
giving out pus' be correct, an emendation of ad galantds to 
sdgalantas would give a good sense : " The barren apacit, 1 
daughter of the black one will fly forth : the boil shall fly 
forth from here, (and) disappear together with the pus." 

ghrtastavas- 

ydsmin deva dmrjata yasmin manusya utd : tdsmin ghrtd- 
stavo mrstva tvdm Ague divam ruha. 12. 2. 17. 

Previous Scholars: * von Schmalze triepf end * BR.,who take -stava- 
to be the stem. Ludwig, p. 480 : ' an dem wisch dich ab und ghrta-triep- 
fend [?], steig empor zum himmel'. Henry: 'puisquo tu d^gouttes de 

beurre, esaui-toi sur lui, | 6 Agni, ot monto au ciel'. " on that 

having wiped off the drops of ghee (?), O Agni do thou mount the sky " 
Whitney, who remarks, " Our mss. seem to read -sla- very plainly [and 
8 PP. reports no variant], but need not prevent our understanding 
instead -and-, if more acceptable ". 

BR. assume a root stu- 4 to drop ', 'to come together ' 
La order to explain stuta- (v.l., sruta) in Hemacandra, -stava-, 
stuka- 'tuft or knot of hair', and stokd- 'drops'. Of these, 
stuta- is only a corruption of sruta- and stokd- has been explained 
by Wackernagel, I. 239. c., as a methathesis of *skota-: &cut- 

1 Apacit- is rightly identified by Bloomfield, p. 504, with apacl- of the 
medical works. The latter seems to be a Prakritisation of the former. 
The commentators of Kau. and AV. also identify it with gandamala. 
BR. give ' scrofulous swelling of the glands of the neck ' for gandamala, 
which is misleading, as the folio whig extract from VSgbhata (Astfinga- 
hrdaya, Uttara. 29, 18) will show : medahsthah kanthamanyakeakaksavath- 
ksanaja malah : savarnan kathinan anigdhan vartakamalakakftin : avaga * 
dhan bahun gandarhd cirapakaibt ca kurvate : pacyante *lparujas tvanye 
sravanty anye 'tikandurah : ndsyanly anye bhavanty anye dirghdkalanu- 
bandhinah: gandamala 'pad seyam durveva kaayavfddhibhak. 



VOL. XVII, PT. I.] INTEBPBETATION OF DOUBTFUL WOBDS. 69 



*to drip'. The remaining stuka- or stu-, as well as 

* tuft or knot of hair ', stavaka- ' bunch ', Beng. thokd ' bunch ', 
contains an element stu~ which gives the idea of a * bunch ' 
or ' bundle '. This, however, would not apply to ghrta. 

On the other hand, if, following Whitney's recommendation, 
we read -snavas 1 , we find -snavas-: snu- 'to drip* as vasas- 

* covering ' : vas- ' to clothe ', -vacas- in suvacas- etc. : vac- ' to 
speak ', vahas- ' offering ' : vah- ' to carry ', pajas- ' brilliance ' : 
pajrd-, etc., Ved. Gr. 126. Then ghrtasnavas- would mean 
"drippings of ghee', object of mrstva, cf. ghrtasnu- 'dripping 
with ghee ? . 

citi- 

devas te cltim avidan brahmana utd virudhah : cltim te vUve 
deva avidan bhumyam ddhi. 2. 9. 4. 

Previous Scholars : Sayana : 'grahavikarad rogina addnam grahadeh 
samvaranam cJiadanam abhistaranam va\ BR. and Weber: 'Sammeln. 
Ludwig, 'pfliickung'. Grill: "Den Gottern und Brahmanen ward's, wie 
man dich Holz zusainmenlegt, Die Gotter all ersahn, wie man zur Erde 
dich zusammenlegt". Bloomfield, p. 292: "The word Mtim is found 

only here and is very problematic we are connecting the word with 

fcinoti in the sense of ' arrange ', ' build up ', having in mind the peculiar 
amulet or remedy dasawiksha * consisting of ten woods ', in st. 7." Whit- 
ney, who renders, 'gathering (?)', remarks, "If it comes from ci, there is 
hardly any other example " . 

Ctti- appears to be connected with cay- l to fear ', ' to be in 
awe of ', 2 thus meaning c apprehension ' and thence ' trouble ' or 
' disease ', just like atanka- which means both ' fear ' and 
' disease'. The hymn is about curing one of a dangerous kind 
of seizure of the joints, said to be caused by demons, and the 
meaning suits quite well : " The gods have noticed thy trouble 9 
the priests, and the plants : all the gods on earth have noticed thy 
trouble ". That the diseased, and not the amulet, is addressed 
is clear from the following verse. 

1 Mss. often confuse at (^) and sn (^) and, in general, t and n, 
cf. atava or slawi in two mss. for snava, Gaastra's ed. of GBr., p. 120, 
n. 2. 

* Cf. Wack., 1. 79. a. : "So von cat/-,, Scheuen" AV. citi- cikihi k\. 
ceki-yate : v. cfiyamana-t'cfiyya cay-ii, sp. weiteres, vgl. gr. rt-/nif." 



70 INTERPRETATION OF DOUBTFUL WORDS. [J.B.O.R.S. 

For formation, of. giti-: gayati, piti-: pa-, payayti (oaus.), 
8'phiti-: sphayate, etc., Wack. I. 79. (a), a, j3. 

jabhya- tardd vdgha- 

hatdm tarddm samankdm akhum A6vina chintdm 6iro dpi 
prstih 6rnitam : ydv&n ned ddan dpi nahyatam mukham dtha 'bha- 
yam krnutam dhanyaya. 6. 50. 1. 

tdrda hai pdtanga hai jabhya ha upakvasa : brahme ' va 
'samsthitam havir dnadanta iman ydvan dhimsanto apodita. 
v, 2. 

tdrddpate vdghdpate trstajambha a 6rnota me : yd aranya 
vyddvara ye ke ca sthd vyddvaras tant sdrvan jambhayamasi . 
v. 3. 

yavatir bhfngd jatvdh kururavo yavatir vdghd vrkaasarpyo 
bdbhuvuh : tdtas tvdm asijyayan, etc. 9. 2. 22. 

Previous Scholars: Sayana: tardafr himsakah (akhuh), jabhya= upa- 
(1 mvakaritvad asmabhir himsya, vaghahavaghnanti avabadhanta iti vaghafr 
patangadayah. Bloomfield and Whitney render tardd- and jabhya- as 
' borer ' and ' grinder 1 respectively. Samatikdm and tipakvasa have been 
left out as doubtful by everybody. Sayana explains the former as aamafica- 
naih bilam sampravisya gacchantam, adj. of akhum, and for the latter he 
reads apakvasah (adagdhah santah). 

Kesava on Kaus. 51. 22 (dealing with AV. 6. 50) has 
samdptam musaka-6alabhMpatanga~tittibJ^ - 

vw-rurU'Salyaka-gosedha-gokrmyMi-svastyayanam ; similar is 
8$yana's remarks in the introduction to this hymn, which is 
directed against the pests to crops. In the well-known verse, 
ativfstir anavrstih 6alabhd musikahSukah; pratyasanna^ ca rdjanah 
sad ete Itayah smrtdh, the most harmful to crops, among animals 
are * locusts', * rats', and 'parrots'. These three must be in- 
tended in vs. 2 and 3 by the words, tdrda : tdrda-pate, * borer ' 
^'rat', pdtanga : vdghapate' locust', cf. Beng. (dial, of Man- 
bhum) baghdulu ' a kind of moth ', Sindhi bagh(g)ai ' a kind of fly 
that sticks to dogs and horses', and jdbhya : tfstajambhah, 
' grinder ' or ' having sharp grinders ' = ' parrot ', which cuts any- 
thing by grinding with its beak. 

The remaining animals in Sayana's and Kesava's enumera- 
tions are included in the words aranya vyddvarah and yi ki ca 
sthd vyddvarah in v. 3. 



VOL. XVII. PT. I.] INTEBPBETATION OF DOUBTFUL WOBDS. 71 

The construction suggests that samankdm in v. 1 is pro- 
bably a verb, second, du., imp., like hat dm, chintdm and fynitam. 
A loss (by ms. corruption) of t , the third member of the conjunct 
(see under kdkutsala-) is probably to be seen after -nk- t samankdm 
thus standing for samanktdm. Now, sdm-anj- has a meaning 
< to consume ', * to devour J (BR., RV. 10. 45. 4 ; 52. 3 : 87. 16) 
which would quite suit the context. 

tlfpakvasa, taken by others as a substantive of doubtful 
significance, also appears to be a verb, probably a corruption 
for -krasa (for a reverse process in the confusion of v and r, see 
ultaradrau). The form *kras- not known to be found else- 
where, may be an enlargement of kram- t to step ' through the 
intermediate krams- which is frequently used. Compare the 
relations of, tarn- ' to be exhausted ', * to perish ': tas-, nam- in 
eved yune yuvatdyo namanta ydd %m u&dnn u&atir ety dccha, RV. 
10. 30. 6 : nams- in ni te namsai plpyane'va y6sd mdrydye y va 
kanya &a6vaeai te, RV. 3. 33. 10 : nas- l to approach with lust', 
6am- 'to kill*: os-, etc., and alternate presence or absence of 
nasal in dan&-:da&- 'to bite/ dans- 'to be worth seeing': 
das- in daamd- etc., nanS-: na6- ' to disappear ', bhran&-: bhrat- 
' to fall ', srans-: sras- ' to fall ', etc. If that be the true reading 
and meaning, *upakrasa would mean ' step nearer ' (that is, in 
order to listen), and would correspond even in this respect (as 
in the names) to a 6rnota me in v. 3. 

tayadardm pdrasvant - 

ydtha pdsas tayadardm vatena sthulabhdm krtdm : yavat 
pdrasvatah pdsas tavat te vardhatam pdsah. 6. 72. 2. yavad 
anginam parasvatam hastinam gardabham ca ydt : yavad d&vasya 
vajinas tavat te vardhatam pdsah. v. 3. 

aydm Indra Vrsakapih pdrasvantam hatdm vidat : asim su- 
nwn ndvam carum ad edhasya 'na acitam vitivasmad tndra uttarah. 
RV. 10. 86. 18. 

I6dnaya pdrasvata alabhate Mitraya gauran Vdrunaya mdhi- 
san, Bfhaspdtaye gavay&s Tvdstra ustran. VS. 24. 28. 

suparnah parjanyo hamso vrko vrsadamfas te aindrd apam 
udro 'ryamno lopa&as simho twikulo vydghras te Mahendraya, 
Kamaya parasvan. KS. V, 7, 11= TS. 5. 5. 21. 



72 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

sa iha kilo va patango va matsyo va Aakunir va simho va 
varaho va para6van (v. 1., -&va) va 6ardulo va puruso va 9 nyo va 
tesu tesu sthanesu pratydjdyate. ankh. Aran. 3. 2. 

Previous Scholars : The commentator simply calls tayadard- (which 
he reads tayodara-) ' a kind of animal ' and no one has yet any suggestion 
as to what it is ; he has nothing to say about pdrasvant- either. Sayarm 
on RV. has: 'he Indra ay am Vfakapih parasvantam parasvam atmano 
visaye 'vartamanarh hatam himsitarh vidat\ On TS. he explains it as 
mahisa, with which may be compared Vj. 65, 21, gavala ca parasvaM ca 
mahisah syad aranyajafr. The commentary to ^fthkh. Aran. calls it ' a 
kind of snake '. BR. suggest that it is probably the ' wild ass '. 

Tayadardm is probably to be read tayad dram, the former 
being the subjunctive of taya-, passive stem of tan- ' to extend.' 
Cf. Whitney, Sans. Gr. 774, " Examples of the transfer of 
stems from the ya- or passive class to the ya- or intransitive 
class were given above (761 b.) ; and it was also pointed out 
that active instead of middle endings are occasionally, even 
in the earlier language, assumed by forms properly passive : 
examples are a dhmayati and vy dprusyat (QB.), Bhuyati (Mai. 
U.)." Or, ta, itself, found in atayi (aor.), tayita (fut.), -taya 
(ger.), tayitr- etc., may be the stem. The verse would then 
mean, ' so that the member may extend sufficiently, made 
big by the wind ', etc. 

From the passages from KS. and Sankh. Aran., quoted 
above, pdrasvant- appears to be a wild animal like the jackal, 
the lion, the tiger, and the boar, and from AV. it is known 
to have a large member like the elephant, the ass, and the 
horse. Both these features are satisfied in its identification 
by Biihler, ZDMG. 48, 63, with Pali palasata-, Asoka, Delhi 
Topra V. 6, Jataka VI, p. 277, v. 1205 and palasada-, Jataka, 
V, p. 406, v. 267 , a which the commentator to Jataka explains 
as khagga * rhinoceros'. Biibler thinks that the intermediate 
steps were * parassanto, * parasato, and Trenckner, Pali 
Miscellany, p. 58, n. 6 (cited by Hultzsch, Asoka Inscrip., 
p. 127, n. 8) considers palasato to be the original of paldsado 



1 Compare with the above passages : tatha panaya- -m- ayanti naria- 
migagana bahu : siha vyaggha varaha ca acchakokataracchayo. palasada 
(v. 1. palasaja, polos ad a, pallasada) ca gavaya mahisa rohlta rurv : eneyya 
varaha c'eva ganino mkasukara, etc. 



VOL. XVIL PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 73 

and the latter, which literally means 'eating leaves', to be 
an etymologizing corruption of the former. 

If this be the true meaning, the proper form is probably 
pdra6vant-, literally, * having an axe (para&u-)\ the reading 
of Sankh. 5ran. Compare, Tchadga-, -dgin-, vdrdhrinasa-, Germ. 
Nasehorn, etc. For syncope of u, cf. para^vadha- ' axe ' (para&u- 
vadha ?), parfoadha- (par6u-vadha ?) ' axe ' : para&u-, par&u-, 
vdnanvat- : * vanana-vat- (Grassmann), indhanvat- : indhana- 
(Ved. Gr. 234), Mmnmaya-, hiranin- : hiranya, etc. 

tiritin- 

yds tvd svdpne nipddyate bhratd bhutva pite'va ca : bajds tan 
sahatdm itdh kKbdrupdms tiritinah. 8. 6. 7. 

Previous Scholars : Sayana : * tiritinah, antardhanena atatdfr.' BR. : 
tirita^' eine Art Kopfputz, Turban, Diadem Arunadatta bei U^^vala, 
die Erklarer zu AK. (3. 6. 3, 30).' Others have followed BR. 

It appears that the meaning head -ornament ' or ' tiara ' 
(Wh.) for tirita- depends solely on a quotation from Arunadatta 
and statements of commentators later than him ; it is not given 
in any of the old Kosas, nor do the modern dictionaries quote 
any occurrence of its actual use in literature. It might have 
arisen from a false reading of kirita-, just as balatanaya-, a 
synonym of Ichadim-, in Amara from that of balapatra- (see 
Ksirasvamin's comment). 

The PTS's Pali Dictionary gives two meanings to this 
word, (1) 'the tree Symplocos racemosa' (lodhra) and (2) 'a 
garment made of its bark.' This second meaning suits the 
passage in question admirably well, for fi the arayas (' niggards ') 
spoken of in this hymn, although vested with spectral qualities, 
are conceived by the poet to have been like a sort of wild 
folk who clad themselves in barks, skins, and tattered clothes. 
There seems to be no point in describing them as wearing a 
head-ornament. Puggalapannatti, p. 51, 'so sanani pi dJidreti 
masanani pi dhdreti chava~dussani pi dhdreti pamsukulani pi 
dharetl tiritdni pi dhdreti ajindni pi dhdreti ', read together with 
v. 11 of this hymn, ye Jcukundhdh kukurabhdh kfttir duriani 
bibhrati; kliba iva pranftyanto vane ye kurvdte ghosam tan ito 
ndtaydmasi (especially the words chava-dussdni , tiritiini and 



74 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.8. 

ajinani in the one and kfttlh and durSani in the other, together 
with tintinah in v. 7), makes it abundantly clear that tirita- 
has the same meaning in both. 



citrani aakdm divi rocanani sariarpani bhuvane javani : 
turmifam sumatim icchdmdno dhani glrbhih saparyami nakam. 
19. 7. 1. 

Previous Scholars: The commentator explains turmitam either as 
turmayo himsakah himsakarinah tan 6yati tanukarotiti turmMa or as turo 
himsakan misati hinastlti turmtia : misa spardhaypm. . . .murdhanyaaya 
talcwyopcyanaS chandasdh. "Seeking favour of the twenty-eightfold (?) 
wondrous ones, shining in the sky together, hasting in the creation (bhuva- 
na), I worship (sapary) with songs the days, the firmament (ri&ka)" 
Whitney, after emendation of turmitam to as^avimMm (or -4ti). 

Tarmi6am seems to be a corruption for * turvi&am ' settle- 
ment of the stars' (cf. tarapatha-, naksatraloka-, etc., for the 
* sky *) the first member (tur) being the genetive singular of 
tr- * star ' and the second vi6- ' settlement.' For interchange of m 
and v, cf. anarmdndm AV. 7.7. 1 : anarvdnam RV. ; amamcw?-: 
amavasi- 'new-moon', Vj. 160, 35; dvradanta in d&rathnan 
drlha dvradanta vilita- RV. 2. 24. 3 prob. dmradanta ( :mrdu-) 
6 the firm became loose, the hard became soft ' (for it Grassmann 
supposes a root mad-, vrand- * to become weak ') ; amatah, 
pardmatah (mss.) for -vatah, Gaastra's GBr. p. 2, fn. 1 ; vangava- 
gadhah, Ait. Aran 2. 1. 1 prob. = -magadhah (Keith); and 
Wack. I. 177. n. For singular giving a plural sense in an 
aluksamasa, cf. Ved. Gr. 275, "A singular case-ending 
(the ace. or inst.) may here indicate a plural sense ; e.g., d&vcm- 
isti- ' seeking horses ', puram-dard- ' destroyer of forts ', 6unesita- 
1 driven by dogs ' (Mna-)". 

Ahani in d is probably elliptical for vi&va dhani. Thus the 
third and fourth pada may be translated : " Desiring good- will 
of the settlement of stars, every day with songs I worship 

the firmament." 

taimatd- 

asitdsya taimatdsya babhror dpodakasya ca: satrasahaaya 
'hdm manydr dva jyam iva dhdnvano vi muncdmi ntitiH iva, 
5. 13. 6. 



VOL. XVH. PT. I.I INTERPRETATION OF DOUBTFUL WORDS. 75 

nir vai ksatrdm ndyati Mnti vdrco 'gnir iva 'rabdho vi 
dunoti sdrvam; yo brahmandm mdnyate dnnam eva ad viadsya 
pibati taimatdsya. 5. 18. 4. 

Previous Scholars : BR. and others consider taimatd- to be the name 
of a kind of snake. Whitney translates 5. 13. 6, " Of the Timatan ( ?) black 
serpent, of the brown, and of the waterless, of the altogether powerful (?) 
I relax the fury, as the bow; I release as it were chariots". To this 
Lanman remarks : " Whitney would doubtless liave revised this carefully. 
The divergences of the translators reflect the uncertainties of the exegesis. 
' I slacken as it were the oars of the wr^th of * etc. Griffith. ' I release 

(thee) from the fury of ' etc. Bloomfield. * Des Asita des Manyu 

Streitwagen gleiohsam spanne [ioh] mir ab ' or * die Streitwagen des 
Grimmes des Asita ' etc. Weber. For d, ' as the string from off (dva) 
the bow'". 

There seems to be a contrast in 5. 13. 6 between taimatd- 
and dpodaka-. The word dpodaka- ' out of water ' means either 
1 not in water* as in RV. 1. 116. 3,. ... naubhir atmanvdfibhir 
antariksaprudbhir dpodakabhih, or 'not containing water', 
'dry', as in AV. 5. 16. 11, where it is obviously used as 
equivalent to drasd- 'sapless'. Taimatd- also appears to mean 
'wet', but modified into 'living in water' at 5. 13. 6 and 
< liquid ' (e.g., the poison of a snake) in 5. 18. 3. It is probably 
a derivative of tema- ' moisture ' with suffix -ta, for which and 
for length of the vowel before the suffix, see under arataK-; 
for superfluous Vrddhi in the first syllable, cf. prainand- 
5. 27. 3 and sauprajastvdm 2, 29. 3, for which other texts, 
including Ppp., have prln- and suprajas-. The first member 
of satrasahasya is obviously the same as satra, which means 
'together', 'altogether', 'all at once', 'always', etc., and 
may here mean 'everywhere'; thus the word would mean 
'prevailing everywhere, i.e., both on land and water'. 
Dhdnvan in d appears to be elliptically used in both senses, 
' bow ' with jyam and ' sandy wasteland ' with rdthan, the 
comparison in the latter case being with the extrication of the 
wheels of a chariot stuck in a sandy piece of land. 

Thus we may translate 5. 13. 4 : " I release (thee) from the 
fury of the wet black snake, the dry tawny snake, and of the 
all-prevailing one, like the string from the bow or ohariots 
(from a sandy tract)." 



76 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

tauld- 

ajyasya paramesthin jatavedas tdnuvafrin: Ague tauldsya 
pra&ana yatudhanan vilapaya. 1. 7. 2. 

Previous Scholars : The commentator derives it from tula- balance ', 
here standing for the sacrificial ladle. Roth and Whitney emend it in 
their edition to taildsya ' of sesame oil ' against all authorities and Ppp., 
which has tulasya. 

Tauldsya is evidently an adjective of ajyasya, and is 
probably from tula- 'a tuft of grass, etc. (here ku&a-). 9 Cf. 
pavitra-, with which ghee was sprinkled. Tula- is used with 
reference to darbhd- at 19. 32. 3. 

duradabhna- 

vilohito adhisthanac chakno vindati gopatim : tdtha va&ayah 
sdmvidyam duradabJina hy iicydse. 12. 4. 4. duradabhnai 'nam 
<l 6aye ydcitam ca nd dltsati : na \smai kamah sdmrdhyante yam 
fidatvd ciktrsati. v. 19. 

Previous Scholars : BR : * Thore tauschend fl.h. durch Schloss und 
Riegel nicht zu halten'. Ludwig : ' unbetrieglich ' at v. 4, but leaves the 
word without rendering at v. 19. Henry, p. 250, sees no reason for, or 
suitability of, tho meaning given by BR. and Ludwig ; ho suggests 
durdaghna- or dur-adaghna- ' killing those who do not give ' as the correct 
reading. " Anaemia (vilohitd) from the station of the dung visits (vid) the 
master of kine ; so is the agreement of the cow; for door -damaging (?) art 
thou called." Whitney (v. 4), who remarks, "Nearly everything in the 

second half verse is doubtful sammdya (given by four of his mss.) 

seems a much more probable form of stem The second person 

ucydse is quite unexpected", and at v. 19, "That the conjectural render- 
ing ('Door-damaging lies she on him' etc.) is extremely unsatisfactory is 
plain ". 

The correct reading is probably duradabhna- ' hard to 
deceive (or harm)', paralleled by ddabdhah * unharmed ' RV. 
1. 173. 1 and duradhdrsa- ' hard to be dared against ' AV. 12. 5. 
17, both epithets of the cow, and nd tti na^anti nd dabhati 
tdskaro nasdm amitro vyathir a dadharsati AV. 4. 21. 3 (=RV. 
6. 28. 3), about the cows. The suffix -na is also found in 
mathnd- 'shaking vehemently' RV, 1. 181. 5 (cf. also 
mathnati and dabhnoti) and after roots like bhanj-, bhug-, rug-, 
pr-, etc., as forming the past participle. 

Samvidyam, as suggested by Whitney, is a better reading in 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 77 

c of v. 4 and in view of the second person ucydse in d, va6ayah 
sdmvidyam seems to be a corruption for va&e 'yah samvidyam , 
' O cow, thou earnest to recognition '. Thus we may translate 
the verse : " Anaemia visits the lord of kine for taking possession 
of the dung ; thus, cow, thou comest to recognition, for thou 
art called ' hard-to-deceive ' ". 

In v. 19 a &- is used in the sense of ' to enter into ' for the 
purpose of doing something evil as in RV. 10. 162. 1, 
brdhmand 'gnih samviddno raksoha bddhatdm itdh : dmivd yds te 
gdrbham durnamd yonim d&dye, and elsewhere. Thus: "The 
1 hard-to-deceive ' gets into him, if he does not wish to give her, 
when asked ", etc. 

dur&d- 

pavdstais tvdpdry akrinan dur6ebhir ajinair utd ; prakrir asl 
tvdm osadhe 'bhrikhdte nd rurupah. 4. 7. 6. 

yt kukundhdh kukurabhdh kfttlr durSani bibhrati ; kliba iva 
pranftyanlo vane ye kurvdte ghosam tan ito ndSayamasi. 8. 6. 
11. 

Previous Scholars : Say ana takes durs&bhih as adjective of ajinaih 
(?=dti8t f a-f6yasambandhibhih) at 4. 7. 6 and reads kftyair chlsyani for kfttlr 
durtfinifti 8. 6. 11. 'Eine Art Gewebe oder Gewand ? , BR. Weber 
on 4. 7. 6, Ind. Stud. 18. 29, follows BR. and remarks, " zu diirQa, Pet. 
W., ist ausser du9ya, dushya vielleicht auch t ft s h a zu vergleichen ", 
but renders it with 'Haute' at 8. 6. 11. (ib., 5, 255). Ludwig leaves the 
word untranslated. Grill (4. 7. 6) : * Fur Decken tauschten sie dich ein, 
Gewebe, Felle gab man her ', etc. Henry (8. 6. 11) : ' Les kukundhas, les 
kukurabhas qui portent des peaux en guise de vetdments ', etc. Bloom- 
field (4. 7. 6) : 'with broomstraw (?), garments ' etc. Whitney renders the 
first durtid- with 'garments' and the second with 'pelts', each followed 
by a query -mark. 

Dur&d- is found at least three times in the Kausika Sutra : 
(i) jihvaya utsddyam aksyoh paristaranam astrhanam hrdayam 
dur$a upanahya tisro rdtrih palpulane vdsayati 11, 16, where the 
tongue and several other parts of a cow rre prescribed to be 
tied up in a dur&a- and soaked in cowdung for three nights ; (ii) 
krmuka$akalam samksudya dur^ajaradajindvakarajvdlena, 28, 2, 
where it is prescribed as an easily inflammable light fuel along 
with worn-out goatskin and broomstraws ; and (Hi) vilumpantdm 
agham iti pan cailam (v.L, &dam, ilam, of which the latter 



78 I NTBBPBBTATION OF DOUBTFUL WORDS. [j.B.O.B.S. 

seems to be the correct reading) durfam vilumpati 85, 22, where 
it is to be spread around (Res'.) a piece of stone (paritilam). In 
all these instances Darila and Kesava 1 render it with jirnaws- 
tra- ' old or tattered cloth '. It will be seen that this meaning 
suits also the two occurrences in the AV. In the first passage 
it is used together with hemp-textures (see pavdsta-) and goat- 
skins, both of which are inexpensive, as a means of exchange 
for a herb. (It may not be out of place to mention here that in 
India even to-day old clothes are very often exchanged for 
herbs from the forest-folk.) In the second a set of wild people 
(or spectres conceived as such, see tiritin-) are said to clothe 
themselves in skin and durSd-. 

Prak. duaa- has two meanings 'cloth' and 'tent', in the 
first of which it must be the same as Sans, dur&d- and in the 
second as Sans, dusya- ' a tent '. Pali dussa- is also to be similar- 
ly assigned. It would appear that in these two languages 
dur6d- has developed a wider meaning, from 'old cloth' to 
' cloth in general '. In Pali chavaduasa- ' a miserable garment ' 
it seems to have retained its original meaning. 

Beng. dhus, Nep. dhusa, Punj. dhussa, and corresponding 
words in other modern Indian vernaculars, all meaning ' a kind 
of woollen covering ', may be connected with it, but the aspira- 
tion is unexplained. 

nyd- dnya- 

drdhd drmhasthiro nyo brdhma viAvasfjo dd&a : nabhim iva 
sarvdtaA cakrdm ucchiste devdtah 6ritah. 11.7 (9). 4. 

1 It may be mentioned 'here that Kedava appears to have flourished 
in the time of King Bhoja of Malwa (o. 1018-1060 A.C.), when the 
Muhammadans were terrorising the country. Speaking about the import- 
ance of dbhicara at the end of Kandika 49, he remarks, pgthivyam dus\a 
utpannah aarvada ca viriatayet : adharmasambhavo due to prajahithsana- 
tatparah : Turueka namna papis^ha devabrahmana-himsakah : ppthivyarfi 
Sfrl-Bhojadeva dharmasarhraksanaya ca : dee tu Malavake utpannah 
Arlrajagfhew ca; again under 47, 22, idam aham Mahmadasya Turns- 
kasya Mutikarnaputrasya pranapanav apayacchami. He was thus a 
contemporary of Uvata, commentator of the VS., and preceded Say an a 
by nearly three centuries. Darila, whose work forms the basis of 
Kea"ava's Paddhati, must have flourished much earlier. 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 79 

a tv ddyd sabardugham huve gayatrdvepasam : tndram dhenum 
sudugham dnyam isam urudharam aramkftam. RV. 8. 1. 10. 

ida hi va upastutim ida vamdsya bhaktdye : upa vo vi6vave> 
daso namaayur & asyksy dnyam iva. RV. 8. 27. 11. 

yo asyd viMjanmana Ue vtevasya ctetatak : dnyesu faiprd- 
dhanvane tdsmai prana ndmo'stu te. AV. 11. 4 (6). 23. 

ydsycte cdtasrah praditiak prthivya ydsyam dnnam krstdyah 
sambabhuvuh : ya bibharti bahudha prandd ijat sa no bhumir gosv 
dpy dnye dadhatu. 12. 1. 4. 

dnyebhyas tva purusebhyo gobhyo dSvebhyas tva : nih kravya- 
dam nudamasi yo agnir jlvitayopanah. 12. 2. 16. 

purusa eve'ddm sdrvam ydd bhutdm ydc ca bhavydm : utd 
'mrtatvdsye *6varo ydd dnyena 'bhavat sahd. 19. 6. 4. 

Previous Scholars : (1) nyd. Say ana : ' dpmha-sthirah dfmhanena 
sthirikfto lokah, nyah netdras tatratyah praninafy\ BR., 'wohl adj. von 
unbekannter Bed.' Grassmann : ' niedersinkend (?) [von ni], enthalten in 
ania.' Henry: 'Affermi et solide toi-m6me, affirmi?', etc. He thinks 
(p. 158) that nyd- may be an artificial formation from anyd- as sura- from 
cwttra-, thus meaning ' no other ', * himself J . Bloomfield, who would read 
dfdhadfmhd(h) and consider nydfy * leader' (sing.): ni- * to lead ': 'He who 
fastens what is firm, the strong, the leader,' etc. Whitney, who considers 
dfihha a verb : ' Being fixed, fix thou, being staunch, nyd ', etc, (2) dnya-. 
Say an a on RV. and AV. 11. 4(6). 23 sees no difference between it and 
anyd- 'other'. His commentary on the 12th book is wanting and at 19. 
6. 4 he reads dnnena (SPP. reads anyena, other texts dnnena). BR : * dnya, 
(3. a + nyo, zusammengezogen aus ni-ya, wie auch zu sprechen ist) f. 
adj. nicht versiegend*. Henry gives a number of substitutes for gd#v dpy 
dnye such as go-sthapdtye, * go-svadhanye, etc. and renders 12. 1. 4d : 
' daigne oette Terre nous accorder la possession des vaches.' At 12. 2. 16 
he as well as Ludwig considers dnyebhyah as equivalent to anyi- and in 
the dative. Whitney : to thee being such, O breath, having a quick bow 
among the unexhausted (Idnya), be homage', 11. 4. 23c,d; 'let that 
earth (blv&mi) set us among kine, also in inexhaustibleness (tdnya), 12. i. 
4d; 'Thee from inexhaustible (tdnya) men, kine, horses * etc, 12. 2. 16. 

The correct reading in 11. 7. 4a appears to be dfdho 
'drmhas sthiro nydh, the loss of a sibilant before a sibilant 
followed by an explosive (as between -drmhas sthiro) being pres- 
cribed by the Pratisakhyas and Katyayana (Wack. I. 287. b) 
and widely followed in the Vedic texts. This gives us a pair of 
contrasted words, drdhdh ( firm ': ddrmhah ' not-firm ' (cf . bhumi- 
dfmhd- 6. 28. 14, 19. 33. 2), sthirdh 'stable': nydh, which 



80 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

should therefor mean 'unstable', 'movable 3 , 'perishable', or 
the like. Nyd- (to be read nid) may easily have this meaning, 
if its relation to root m- in the sense, 'to take away', 'to 
remove ', is considered to be the same as that of priyd- ' dear ' 
to the root pri- ' to please '. 

Anya- would thus mean ' immoveable ', ' imperishable ', and 
thus be an exact equivalent of amfta-, which is sometimes an 
adjective, meaning ' imperishable ', and sometimes a noun, mean- 
ing ' nectar ', often applied to sacrificial butter or the soma* 
drink (see Grassmami, s.v.). At 12. 1. 4 and 12. 2. 16 dnya- is 
probably used in the sense of sacrificial butter ; at RV. 8. 1. 10 
and 8. 27. 11 it qualifies isam 'libation', understood in the 
second passage; and at AV. 19. 6. 4 it means 'nectar', thus 
giving a much better sense than the other texts which read 
dnnena. Anyeau at AV. 11. 4. 23 is probably a vocative of 
dnyesu : ' O one of inexhaustible arrows, to thee of quick bow, 
O Prana, be homage.' 

patu(tau)rd- 

pratighndnah sdih dhdvantu 'rakpaturav aghnanak : aghari- 
mr vikeSyo rudaty&h puruse hat6 radite Arvude tdva. 11. 9 
(11). 14. 

Previous Scholars : Sayana : ' urah vakaahsthalam paturau tatpradeSau 
ca\ Bohtlingk: 'ein bestimmter Korperteil*. Bloomfield who renders 
with 'thigh', remarks [Reprint of A.J.P. XI. 3 (pp. 319-366), p, 22]: 
** Our translation of the obscure word is based upon Kaug. 84, 10, trih 
prasavyam praklrnakeyyah pariyanti daksinam urum aghnanah. The 
word seems to contain uru with some modifying adjective, perhaps pfthu 
in a Prakrtic form." Henry suggests emendation to *patair avaghnann\i 
and renders : ' Qu'elles accourerit, celles qui se meurtrissont de coups, | 
arrachant les voiles qui couvrent leurs seins,' etc. Whitney notes : " I 
follow both translators in rendering pa\aurd by ' thigh ', although it is not 
too acceptable, considering the familiarity of uru as name for '.thigh ' ". 

Paturd- or -taurd- is evidently identical with patora- in K8. 
V. 13, 11-12, Agneh paksatih Sarasvatya nipakaatih Somasya 

trtiya devdnam dvada&i dyavaprthivyoh par&vam Yama- 

sya paiorah, and patura- in a corresponding passage in TS, 5. 7. 
21-22. This is about the ribs of a horse. VS. 25, 4-6 and MS. 3 . 
15. 21-22 in the corresponding passages count thirteen ribs and 
leave out patura- . Caraka, Sarlra. 7,5, counts only twelve ribs on 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 8i 

each side, while GBr. 1. 5. 3, thirteen. It thus appears that there 
have been two views about the exact number of ribs, and in 
any case patura- seems to be outside, but just following, them on 
each side. It is most probably the bone, on each side, just about 
the thigh-joint and below the waist. VaijayantI, p. 22, gives 
patura- as a name for the fourteenth lunar night. . . . pratipat tv 
ekapaksatih : paksatis ca 'tha paturo bhutesta ca caturda&i : ni- 
paksatir dvitlya syat. This also supports the above contention, 
the two fortnights in a month being considered its two sides 
(paksa). 

In the above passage paturau evidently stands not actually 
for these bones, but for the two sides of the belly just near 
them, referring to the common phenomenon of the striking 
of the breast and the belly by women in grief. 

parusd- 

darbhdh Sods tarunakam d&vasya varah parusdsya varah : 
rdthasya bdndhuram. 10. 4. 2. 

Previous Scholars : Ludwig : " das junge darbhagras 1st flamme [der 
schlange], der pferdes scheif, des paruSa schweif [1st flamme fur die 
schlange], | [ebenso] des wagens sitz." Henry : " Ardour brulante [contra 
le venin] eat le brin de darbha, ainsi que le jeune brin [d'orge ?], | lecrin de 

cheval, le poil d'homme|| le caisson du char " Bloomfield 

renders paruad- with ' the shaggy one ', but remarks (p. 607), " For the 
unintelligible parushasya we are tempted to substitute arushasya relying 
upon the oft-emphasised whiteness (sveta) of Pedu's horse". Whitney : 
"Darbfoa-grass, brightness, young shoot (? tartinaka) ; horse's tail-tut r 
rough one's tail-tuft ; chariot's seat (? bdndhura)." 

Parusd- here probably refers to the bull. It is used as an 
epithet of uksdn- < bull ' in RV. 5. 27. 5, ydsya ma parusah Satdm 
uddharsdyanty uksdnah, and of gau- c bull ' or ' cow ' in RV. 6. 27. 
5, utadah paruse gdvi sura& cakrdm hiranydyam: ny alrayad 
rathitamah, and 8. 93. 13, tvdm etdd adharaydh krsnasu rohimsu 
ca, : pdrusmsu rfaat pdyah. This epithet is due probably to its 
shagginess (of the tail ? Cf. puruvara- 'having an', ample tail- 
tuft', another epithet, RV. 1. 139. 10), as it is applied elsewhere 
also to urna- e wool ' (RV. 4. 22. 2, see Grassmann, s.v. parusd-). 

Parusa- seems to be the correct reading for purusa- (note 
that two of Wh.'s and one of SPP.'s mss. read purusa- for 



82 INTERPRETATION OF DOUBTFUL WOBDS. [j.B.O.R.S. 

parusa- in the above passage) in AV. 6. 38. 4, rajanye dundnbhav 
ayatayam d6vasya vaje purusasya mdyau: tndram ya devi 
subndgd jajana sa na aitu vdrcasd samvidana, and 19. 49. 4, 
simhdaya ratry u6ati plmsdsya vyaghrdsya dvipino vdrca adade : 
d&vasya vradhndm purusasya mdyum puru rupani Icrnuse vibhaii, 
as mdyu- 'bellowing', 'lowing', etc., is usually applied to an 
animal such as the cow or the sheep, cf . Whitney's remark at 
6. 38. 4, " Mayu is not properly used of purusa c man ', and the 
expression is obscure and doubtful ". 

pdrsana- 

Indrd-Soma vartdyatam divas pdry agnitdptebhir yuvdm 
dsmdhanmabhih : tdpurvadhebhir ajdrebhir atrino ni pdr&ane 
vidhyatam ydntu nisvardm. 8. 4. 5=RV. 7. 104. 5. 

girdyaS tin ni jihate pdr&anaso mdnyamanah : pdrvatax cin 
ni yemire. RV. 8. 7. 34. 

ydd vllav Indra ydt sthire ydt pdrSane pdrabhrtam : msu 
spdrhdm tad a bhara. RV. 8. 45. 41. 

Previous Scholars : Sayana : parsvasthipradese, AV. =parsvasthane at 
RV. 7. 104. 5, pidyamanah or sprAyamariah at RV. 8. 7. 34 ; and vimar- 

&ana-ksame at 8. 45. 41. BR.: 'm. Einsenkraig, Abgrund, Kluft 

Nach Naigh. 1. 10. Wolke, wohl derselben ursprungs wie parSuS 

In RV. 8. 45. 41 pdrana- is contrasted with vidu- ( strong ' 
and sthird- ' firm ', while AV. 8. 4. 5, when compared with 8. 4. 
17d, tdm pratydncam arcisa vidhya mdrmani, and 8. 6. 24d, 
hrdaye 9 dhi ni vidhyatam, would point to some 4 tender ' part of 
the body as its meaning. The meaning 'soft', 'tender', not 
only suits these two passages but also the third. 

The original meaning seems to have been what Sayana 
gives at RV. 8. 75. 41, viz., vimar6anaksama- * worthy of being 
touched or felt ' : spara- l touch ' with suffix -ana. This suffix 
-ana, used to form adjectives, is found in several words : 
dpnav-ana-: *apnu(?), urdhvas-and-: urdhvd(s)~, jrayas-and-: 
jrdyas-, dhiyas-and-: *dhiyas- 9 pftkav-ana-: prihu- (see under 
su&ima-), mandas-and-: *mandas- 9 rabhas-and-: rdbhas-, vasdv- 
ana-: vdsu-, vrdhas-and-: vrdhds-,6ava8-and~: &dvas-, sahas-and-: 
sands-, etc. Grassmann and Macdonell (in Vedic Grammar) 
consider it identical with the participial suffix, but added to 



VOL. XVII. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 83 

a double stem of the root, in the case of jmyasdnd-, dhiyasand- , 
mandasand-, rabhasdnd- etc. It seems however more likely to be 
a secondary suffix in view of the above examples, placed side by 
side with the corresponding nouns. 

pavdsta- 

pavdstais tvd pdry akrlnan dur&ebhir ajinair utd : prakrir asi 
tvdm osadhi 'bhrikhdte nd rurupah. 4. 7. 6. 

dbhur v auksir vyu ayur anad ddrsan nu purvo dparo nu 
darsat : dve pavdste pdri tdm nd bhuto yo asyd part rdjaso vivesa. 
RV. 10. 27. 7. 

Previous Scholars : Sayana : pavastaih pavanaya astaih sammarjani- 

trnaihj on AV. and * pavaste pavatir gatikarma mahatvena sarva- 

syabhibtutvanaya gacchantyau (dyavaprthivyau),' on RV. BR. : * Zeltdecke 
oder dergl. AV. 4. 7. 6. du. bildlich von Himmel und Erde RV. 10. 27. 7.' 
Oldenberg, RV. II. p. 227, remarks : " Pavdsta dunkel. Ungewissheit der 
auf kombination von AV. IV. 7. 6. und KauS. 28. 2. beruhenden Gleich- 
setzung mit avakarahsti schoii Bloomfield S. B. E. 42. 377f. hervorgehoben ; 
eberiso die Schwiorigkeit, auf diesem Wege Verstandniss unseres st. 
zu fordern. Unzutreffend Weber, Ind. Stud. 18. 29." 

Pali potlhaka- (' a cloth made of makaci fibre ', PTS's P.D.) 
is the regular representative of Sans. pavasta(ka) and, as we 
shall presently see, its meaning suits both the occurrences of 
the latter. In Jataka IV, p. 251, potthakam is explained by 
ghana- or sana-satakam. From the descriptions of pottltaka- 
given in Anguttaranikaya p. 246 and Puggalapannatti p. 33, 
sanasataka-, ' a texture made of hemp ', appears to be the 
true reading. It is described there as having an indifferent 
colour (dubbanno), rough (dukkhasamphasso), and cheap (appag- 
gho) 1 which are exactly the characteristics of a hemp-texture. 
It would appear that a rough texture of this kind was used 
as a wrapper just like a blanket, cf. Kaus. 57. 13, sarvesam 
ksauma&anakambalavastram, and Puggalapannatti p. 51, so 
sdndni pi dhdreti masdnani pi dhdreti, etc. 



1 tayo potthaka : navo pi potthako dubbanno c'eva hoti dukklw- 
saniphasso ca appaggho ca, majjhimo pi potthako dubbanno c'eva hoti 
dukkhasamphasso ca appaggho ca, jinno pi potthako dubbanno c'eva hoti 
dukkhasamphasso ca appaggho ca : jinnaih pi pottliakam iikkJwUiparimaj- 
janam va karenti sankarakute va nam chaddenti. Pug. 33. 



84 INTERPRETATION OF DOUBTFUTi WORDS. [j.B.O.R.S. 

Accordingly, the RV. verse may be translated: "Thou 
becamest, thou grewest up, thou didst attain age; now the 
first one, now the other one has pierced. Two hemp-wrappers 
(i.e., the heaven and the earth) wrap, as it were, him who 
has spread beyond this region." And the AV. one, "They 
bought thee for hemp- wrappers, for tattered cloth and for 
goat-skin ", etc. 

Av. past- 'skin of a man', Pers. post 'skin', later Sans. 
piAStaka- and Pali potihaka- 'a book', form quite a different 
group of words. 

Bloomfield's (SEE. 42, 377 b.) and apparently Sayana's 
(AV. 4. 7. 6) identifications of pavdsta- with avakara-, on the 
ground that the latter word is found with dur&a- and jarad- 
ajina- in Kaus. 28. 2, is without any force, since the words 
are used in different contexts. Pavdsta-, durd~, and ajina- 
iu AV. are referred to as inexpensive means of exchange, 
while dura-, jarad-ajina-, and avakara- in Kaus. are prescribed 
as easily lighted fuels. 

prsti- 

Previoua Scholars: Sayana :=par6u- RV. 10. 87. 10, AV. 2. 7. 5, 
4. 3. 5, =par6vavayava~ AV. 2. 32. 2, p$8thavam6asya abhito vartamanuh 
partuh 11. 10. H. BR. : 'rippe (=ptir$u)\ Grill, Henry, Whitney, and 
other scholars generally follow BR., except when pgsti- and pdrsu- come 
together in the same sentence and where they try to differentiate with 
4 side ' (' flanks ', Henry) and * rib ' (* cote ', Henry) respectively. Uhlen- 
beck, Etym. Wort., has: "pr stfs f. rippe, vgl. afgh. pititai rippe and 
par^us rippe, sichel. AV. parStis np. puSt rticken gehoren nicht hierher, 
sondem zu pfstham. " Pfsti denotes not *rib', which is partu, but a 
transverse process of a vertebra, and so the vertebra itself", Ved. Ind., 
. v. 8arira. This last one will appear to be the right meaning. 

It would appear from the above that scholars are often 
doubtful as to the precise meaning of prsti-, especially when 
they have to distinguish between it and pdr&u-. Prsti- is 
distinguished from pdr&u- in AV. 9. 7. 6, devanam pdtriih prstdya 
upasddah pdrSavah, and 10. 9. 20, yak prstir yaca pdr&avah ; 
and from par6vd- in 9. 1. 34, yah parSvi uparsdnty anuniksanti 
prsfih. In 12. 1. 34, uttanas tva praticlm ydt prstibhir adhi 
Semahe (addressed to the earth), men are said to lie on the 
earth, facing upwards, with their prstis. Whitney and Griffith 
here give a derivative meaning 'stretched out' to uttand-, 



VOL. XVH. PT. I.] INTEBPEETATION OF DOUBTFUL WOBDS. 85 

apparently in order to be consistent with the meaning * rib '. 
But uttdnd- usually means, not only in classical but also in 
Vedic Sanskrit, ' lying on one's back ', ' facing upwards ', of. 
RV. 10.142.5, bdhu ydd Agne anumdrmrjdno ny&nn uttdnam 
anvesi bhumim, and 10. 27. 13, etc., where uttdnd- is contrasted 
with nydn ' facing downwards '. In 18. 4. 10 horses are called 
prstivahak, which Whitney renders f back-carrying (?)'. These 
instances show that prsti- is not the same as pdrsu-, but is 
either the back itself or some part of the body at the back. 
GBr. I. 5. 3, just after counting the pdrsus, analyses the spine 
(anuka-) thus : anukam trayastrimso, dvatrlmsatir hy evai 9 tasya 
prstikundilani bhavanty, anukam trayastrimsam, tasmdd anukam 
trayastrimsah, where it (viz., the spine) is said to have thirty- 
two rings (?) of prsti- [cf. Susruta, arlra. ch. 5., prsthe trimsat 
(asthmi)]. The number and the description agree with those of 
the transverse processes of the vertebra. 

From this the meaning was easily widened to indicate 
the spine or the back itself, as is seen in the case of prstivahah. 
Hastyayurveda of Palakapya, which seems to be a fairly 
ancient work, uses at 3. 9. 55 prstau for prsthe. A similar use is 
also found in Kalyanamandirastotra, Ind. Stud. 14, 38t>, 
and Simhasanadvattrimsika and Pancadandachatraprabandha, 
ib., 15, 378. 

In the light of the above Pali pitthi and pitthl (f.), PrSk. 
patthi, pitthi, and putlhi (f.), Guj, puth (f.), Sindhi puthi, puthi, 
Mar. path, and H. pith (f .), all meaning ' back ', are to be traced 
back to prsti-. Cf. Turner, Dictionary of Nepali, s.v. pith and 
Bulletin of the School of Oriental Studies, vol. V. p. 124, s.v. 
pustai. It is also clear now that Av. parstis and nP. puSt do 
really belong to prsti-, not to prstha- as Uhlenbeck thought. 

pestra- 

ydt te ristdm ydt te dyuttdm dstl pestram ta atmdni : dhdta 
tdd bhadrdyd punah sdm dadhat pdrusd pdruh. 4. 12. 2. 

2/6 nah sdpdd dsapatah sdpato yds ca nah sdpdt : sune pes- 
tram iva 'vaksdmam tdm prdty asydmi mrtydve. 6. 37. 3. 

Previous Scholars : Say an a roads pretham (^=priyatamam) at 4. 12. 2 

and pest am (piatam) at 6. 37. 3. * Knochen vielleicht von 1. pi6 \ 

BR., followed by Ludwig, Grill, and Griffith. Weber, Ind. Stud, 18, 47, 



S6 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

renders with ' Quetschung ', and Hillebrandt, Ved. Chrest., vocabulary, 
remarks: "Ich stelle das Wort zum Wurzel pis, zermalmen, und ver- 
muthe: losgeschlagenes Stuck Floisch, Fleischf etzen ". Bloomfiold also 
renders with *bone% although not satisfied. Whitney renders the two 
passages respectively : " What of thee is torn (n'p), what of thee is in- 
flamed (?dyut) 9 is crushed (?p&8tra) in thyself may Dhatar excellently put 
that together again, joint with joint", and, "Whoever shall curse us not 
cursing and whoever shall curse us cursing, him, withered (?), I cast forth 
for death, as a bone (??) for a dog ". 

Bloomfield, SBE. 42, 385, has already suggested a connec- 
tion of pestra- with pi&tid- and pe&i-, and that, like the other 
two, it may also mean f flesh '. In fact all these three words 
contain the same root pi&- ' to prepare, make ready (especially 
meat, by cutting it up and carving it) ', followed by different 
suffixes, -tra (Ved. Gr. 152), 4a (ib. 145), and 4 (ib. 131) 
respectively ; cf . janik and a-janih, jatdm, janitram RV. 7. 34. 2, 
56. 2 (Say.), 'birth': jan- 'to be born', nrtik, nrttdm 'dance': 
nrt- ' to dance ', vasitam, vaslram ' cloth ': vas- to clothe ', etc. 
For root pi- in this sense, cf. AV. 12. 5. 36, Sarvdh krudddh 
pUydmana Simida pi&itti (brahmagam). The word pi&itd- itself 
occurs twice in the AV., at 5. 19. 5, krurdm asya a^dsanam 
trstdm piitdm a&yate, and at 6. 127. 1, vidradhdsya balasasya 
lohitasya vanaspate : visdlpakasyau 'mdhe mo'chisah pisitdm cand, 
where it seems to have a figurative meaning. 

This meaning not only suits both the contexts but seems to 
be required by them. AV. 4. 12., where the first passage 
occurs, is all about the healing of fractured bones, torn and 
damaged flesh, etc., cf. v. 3, sdm te mamsdsya visrastam sdm 
dsthy dpi rohatu, and v. 4, dsrk te dsthi rohatu mamsdm mamsena 
rohatu. In the second passage, a piece of meat offered to a dog 
would make a better simile than a bone for a man offered to 
death, for in the former case there is also the suggestion of 
being completely devoured. 

pr&rtha- 

anyaksetre nd ramase vai sdn mrdaydsi nah: dbhud u prar- 
thas takma sd gamisyati bdlhikan. 5. 22. 9. 

Previous Scholars : " Lies prd-arthas, nach PW. : ^> Ausriistung zur 
Reise<^; vielleicht aber praedic. zu to/kmd=sfortstrebend (vgl. arthay+pra), 
Ludwig : ^>begierig nach dor Ferne<^ Hillebrandt : 



VOL. XVII. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 87 

Grill. '-The pada-reading in c is pra-drthah ; prd-ar~ would better suit 
the meaning given '.ready to set out ', lit. * having an object in front ' ". 
Whitney. 

Prartha- seems to be a corruption for prarthya-, ' amenable 
to supplication ', cf. RV. 1. 82. lc,d, yada nah sunfiavatah 
kdra ad arthdydsa id yoja nv tndra te hdri, and Grassmann, 
s.v. arthay-, meaning (2). This (prarthia-) is required also by 
the metre and would save us an awkward resolution and a forced 
meaning. The loss of y is only anotber instance of the tenden- 
cy of the AV. mss. to drop the last of a conjunct of three, see 
instances under kdkutsala-, to which may be added aks(y)au 
4. 3. 3 (majority of Wh.'s mss.), tdlp(y)ani 14. 2. 41d, meks(y)~ 
dmi 1. 102. 1, rdks(y)amanah 18. 4. 70, etc. 

bhrmald- 

yds te sarpo vfcikas trstddamSma hemantdjabdho bhrmalo 
guha &dye : kfmir jinvat prthivi ydd yad ejati pravfsi tan nah 
sdrpan mo'pasrpad ydc chivdm tena no mrda. 12. 1. 46. 

Previous Scholars : The commentary is wanting oil this kSnda. BR : 
* adj. belaubt, torpidus '. Wackernagel, I. 63.y.: ' v.: bhrmu- ,,Verirrung' fr 
bhfmi-, bhfmi-, AV. bhrmald : v. bhramd- ,,Lohe" sp. bhramati ,,schwei- 
fen " '. Ludwig, p. 548 : '* der earpa, der stechende, mit hartem zahn, der 
dir von winter erfroren, der sich hin und her schlagehide Im verborgen ligt *\ 
Bloomfield : ' The serpent, the scorpion with thirsty fangs, that hiber- 
nating torpidly lies upon thee ' etc. Henry : * Tori serpent, ton scorpion a 
Tapre morsure, | qui englouti par Fhiver git assoupi dans sa cachette* etc. 
Whitney : " What stinging (vfScika) harsh biting serpent of thine lies in 
secret, winter-harmed, torpid (?bhfmald); whatever worm, O earth, becom- 
ing lively stirs in the early rainy season let that crawling not crawl upon, 
us; be thou gracious to us with that which is propitious". 

The Ppp. reading bhramalo suggests that our bhrmald- may 
be identical with bhmmara- 'a humble bee'. Confusion of ra 
and r in mss. is very common ; cf. for instance, prsti- orprsthi- 
in some mss. for prasti- AV. 10. 8. 8, 13. 1. 21, recorded by 
Whitney in his translation ; kramuka-, v. l.toTkrmuka- Kaus. 28. 
2 ; pra&nyam iorpr&nyam in DaruVs comm. to Kaus. 11. 16 ; drati- 
tordrti' Kaus. 38. 12 ; ms. Ech hrdam, PBh hrdayam iorhradam 
ib. 52. 5, etc. Kesava commenting on ynUayor ma no deva ya# 
te sarpa (the present verse) iti SayanaSalorrarak parilikhctfi, 
Kaus. 51. 17, remark?, aiha. sarpadisvastyayanam ucyate: sarpa- 



88 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 



^ ; etesam 

bhayam na bhavati, and again under the following sutra, ........ 

ahibhaye vrfoikabhaye maakabhaye bhramarasamghe krmibhaye 
etc. His enumerations, especially the second one, when equated 
in order of presentation with our verse, show that according to 
him trstddam6ma= dvidam&a-ma&aka- or ma$aka- and bhrmald- 
= bhramara-. For trstddam&md cf . also AV. 7. 56. 3, urbhdsya 
trpradam&ino madkasyd 'rasdm visdm, and Sans, dam&a-, Beng. 
etc. d&& ' a gnat '. This makes it clear that in Kesava's time 
(c. llth century) either the reading itself was bhramard- or 
bhramald- instead of bhrmald-, or the latter was understood to 
be identical with the former. The above verse may accordingly 
be translated: " The serpent, the scorpion, the harsh-biting 
(mosquito), the bee of thee, that, pressed (lit., ' ground' : jabh- 
'to grind') by winter lie in concealment; whatever worm, 
being lively, O earth, stirs up in the early rainy season, let that 
crawling not crawl on to us, favour us with that which is pro- 

pitious ". 

man(n)au 

deva imam mddhuna sdmyutam ydvam Sdrasvatydm ddlii 
manav acarkrsuh : tndra dsit sirapatih Satdkratuh fona&a dsan 
Marutah sudtinavah. 0. 30. I 1 . V.I., ritan&v-. 

ni tigmdm abhy dm&um sidad dhotd manav ddhi : jusdno 
asya sakhydm. RV. 8. 72. 2. 

dyukta sura etaam pdvamdno manftv ddhi : antdriksena 
yatave. ib. 9. 63. 8. 

raja medhabhir tyate pdvamdno matiav ddhi: antdriksena 
yatave. ib. 9. 65. 16. 

Previous Scholars : All have taken manau (or -nau) as if identical with 
mdnau ( man' or * Manu * (loc.), although some, like Whitney, are not quite 
satisfied. 

The accent indicates that man(n)au may be altogether 
different from mdnu-; and man(n)Uv ddhi with verbs like acarkr- 
suh ' ploughed', sidat ' sat ', pdvamanah : being purified ', suggests 



1 "It occurs also in TB. (ii. 4. 87; exactly repeated in AP. QS. VI. 
30. 20; PGS. iii. 6. 1.), MB. (ii. 1. 16), and K. (xiii. 15). The TB. version 
begins with etdm u tydm mddh- (HO MB. also), and it gives in b sdrasvatyas 
and man&v '\ Whitney . 



VOL. XVH. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 89 

that it may refer to some place or plot of land on which these 
actions could be performed. Now, Beng. has a word mdna 
( < * manyd ? ) which means ' a fertile piece of land on either 
side of a river ' (so near that it is often flooded when the river is 
full). Cf . also mani-rudhaka-, synonym of 6imbidhdnya- (podded 
grain, such as pulses) Kalpadrukosa p. 142, 89, timbidhanyam 
jalaklinnam sdnkuram manirudhakam. The application of 
such a meaning to the first verse, where the river Sarasvatl (cf. 
TB. reading -svatyas) is mentioned, is clear, while the other 
three verses relate to Soma, who is frequently connected with 
the river or river-bank ; cf . in the same hymns, duhdnti saptai 
'kdm upa dva pdnca srjatah : tirthe sindhor ddhi svare, 8. 72. 7 ; 
tdm im mrjanty dydvo hdrim nadisu vdjinam : indum Lndrdya 
matsardm, 9. 63. 17 ; yd drjikesu kftvasu ye, mddhye pastyUndm : 
ye vd jdnesu pancdsu (somdsah aunvire), 9. 05. 23; also 9. 76. 1, 

8. 96. 14, 15, etc. l 

madhyama&ir 

ydsyd ' njana pramrpasy dngam angam pdrusparuh : tdto 
ydksmam vi bddhasa ugro madhyama&ir iva. 4. 9. 4= RV. 10. 97. 
12, VS. 12. 86 with v.l.. ydsyau 'sadhih prasdrpatha in a, and 
bddhadhva in c. 

Previous 1 Scholars : Say ana : * madhyame antariksasthatie Sete samcara- 
tlti madhyamalh vayuh\ or, * arir mitram arer mitram iti niti^astroktct- 
rajamandalamadhyavart% raja ' (AV.), and, ' madhyamasthane vartamano 
raja* (RV.). BR. : *etwa intercessor '. Grassmann : *wol; derin derMitte 
aich lagemde als Bezeichnung etwa des Heer-f uhrers '. Whitney : 
*' Madhyamagl is of obscure meaning; 'arbiter', as conjectured by BR. 
seems very implausible [ BR. express their conjectural meaning by the 
Latin word intercessor ; by which, I suspect, they intend, not * mediator ', 
but rather ' adversary ' or * preventer ' of the disease, which would be 
plausible enough ] ; more probably ' midmost man ', like madhyame sthfi or 
chief (see under iii. 8. 2.), and madhyamaqi used especially of the leader 
about whom his men encamp, for his greater safety, in the night '". 

In the simile in the second half we have on one side, 
* thou drivest away disease ' a subject, an object, and a 
verb , while on the other side apparently only the subject 
(madhyamaSir) and the verb (the same vi-bddh-). This suggests 

1 Cf. Punj. mand * low moist ground on the bank of a river or 
stream % which cannot in the ordinary way come from the same source as 
Beng. mana, and Telugu mann * earth, soil '. 



1)0 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

that we have probably to split up madhyamaSir into two words, 
mddhyam, ace. sing, of mddhya- e the central part', and ofr, 
nom. sing, of *a$ir (for accent and gender cf. vandhur-), 
probably * a boring instrument ' or ( awl ' as will appear from 
the following correspondences. Walde, Idg. Wort., s.v. a 
(z. T. auch ak-) ,,scharf, spitz, kantig; Stein ", gives a detailed 
account of words formed with this element, of which those 
formed with -I (p. 29) seem to be identical (at least some of 
them) with our *a&ir-: " arm. asetn .,Nadel", aksl. Os(l))la 9 
Wetzstein", ags. egle pi. Ahrenspitzen, Grannen", engl. ails 
,, Gra.nnen ", ahd. ahil m. Ahren-spitze, Granne ", nhd. (dial.) 
agel, tigcl yyfe&tnca," und Achel ; acymr. ocolin ,,cos", ncymr. 
ogolen, und (h)ogalen, nbret. higolen Wetzstein" ; lat. aculeus 
Stachel " ; anord. soft-all Fleiachgabel " ; cymr. ebill 
.,Bohrer" corn, epill Jtoern .,clav", mbret. ebil ,,Pflock, 
Stiff, Nagel <c ." 

Assuming this modification of the text and meaning of 
*o&>-, Whitney's translation may be modified: "Of whom- 
soever, O ointment, thou creepest over limb after limb, joint 
after joint, from thence thou drivest away the ydksma 4 as 
a sharp boring instrument the middle portion (of anything, 
by cutting a hole) ' ". 

mura mumdeva- 

1 . mum- 

ya sasApa sdpanena yti 9 yMm mumm adadhe : ya rdsasya 
hdranaya jatdm arebM tokdm atlu sa. 1. 28. 3=4. 17. 3. 

Mura- has been rightly identified with mula- ' root ', but the 
import of c seems to have been misunderstood. Thus the 
commentator takes muram as adj. of aghdm, meaning either 
adibhutam or murchakaram, and reads adade (=parijagraha y 

krtavati 'tyarthah) for a dadhe ; Grill, ' hat gift'ge Wurzel 

angewandt * ; Whitney, ' . . that hath taken malignity as her 
root (? mura). 9 As, however, a dha- usually means 'to lay', 
'to deposit', 'to bury or impregnate ', cf. RV. 9. 19. 5, kuvid 
vrsanydntibhyah puruino gdrbham adddJiat, and roots are men- 
tioned as having been used for sorcery, cf. mulakfi- in ydh 
krtyakfn tnulakfd yatudhfi.no 4. 28. 6, and mulin- in krtyakftam 



VOL. XVII. PT, I.] INTERPRETATION OF DOUBTFUL WORDS. 91 

valaginam mulinam fopatheyydm 5. 18. 12, the rendering should 
rather be, c that hath buried the wretched root*. Cf. 5. 18. 8, 
yam te krtyam K&pe 'vadadhuh 6ma6dne va nicakhnuh y or 10, 1. 18, 
yam te barhisi yam 6ma6ane ksetre krtyam valagdm va nicakhnuh. 

2. muradeva- 

dyodamstro arcisd yatudhandn upa spr&a jdtavedah sdmid- 
dhah : cijihvdyd muradevan rabhasva kravy&do vrstva 9 pi dhatsvd 
'sdn. 8. 3. 2=RV. 10. 87. 2. 

pdrd Srnihi tdpasd yatudhandn pdrd 9 gne rdkso hdrasd 
Srnihi : pdrd 'rcisd muradevan chrmhi pdrd 'sutfpo abhi $66ucd- 
nah. 8. 3. 13 ( = 10. 5. 49)=RV. 10. 87. 14. 

Indra jahi pumdmsam ydtudhanam utd striyam mdydyd 
&a&addndm : vignvdso muradeva rdantu ma te dr$ant suryam 
uccdrantam. S. 4. 24= RV. 8. 104. 24. 

Previous Scholars: Sayana: =maranakrida raksasah (8. 104. 24), 
mudltadevan marakavyaparan (10. 8. 2), maravyaparan (ib. v. 24) on RV., 
and, mulena ausadhena dlvyanti, or mudhah santo dwyanti te (8. 3. 2), 
maranoia kannana dlvyanfiti (v. 13), maranakrida mulena visaiisadhya 
dlvyantUi va (8. 4. 24) on AV. BR : *=muladeva KaQ, zu P. 8, 2, 18, Vartt. 
2, Bez. gewisser Unliolde.' Grassmann : Bezeichnung damonischer Wesen 
oder ihrer Verehrer. ' Whitney : * falseworshippers.' Henry renders 
wifch 'demons' or 'impies', but remarks in his commentary to 8. 3. 2 
that it is probably a possessive compound with m&ra- * root ', and not 
murd- ' foolish *, as the accent on the first member shows. 

As Henry has already pointed out, the accentuation shows 
that the first member in muradeva- is mura- ' root ', thus the 
word meaning, c having the roots as one's god '. In view of 
the importance of the use of roots in sorcery, as seen under 
mura-, this implies simply that the sorcerers were so called 
because they were devoted to and expected everything from the 
roots just as much as the Rsis did with regard to the gods. 
That roots were actually worshipped as fetish and their wor- 
shippers were later confused with demons, as Henry would 
suggest, does not seem to be a necessary implication. 

ma- 

Vends tat pa&yat paramdm guhd ydd ydtra vi&vam bhdvaty 
ekarupam : iddm pfnir aduhaj jayamdndh svarvido abhydnusata- 
vrak. 2. 1. I. 



92 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

sdnn uchiste dsam* co 'bhau mrtyur vUjah prajapatih : lokya 
ucchista Ayatta vrd ca drd$ ca 9 pi Mr mdyi. 11. 7 (9). 3. 

stdmbhtd dha dyam sd dharunam prasUydd rbhur vajaya 
drdvinam ndro goh : ami svajam mahisdS caksata vram menam 
dtwsyapdri matdram goh. RV. 1. 121. 2. 

f'^vdsa svdsre jyayasyai yonim aralg dpaity asyah praticdks- 
y&v: vyucchdnti ra&mibhih suryasya 9 njy ankte samanaga iva 
vrah. Ib. 1. 124. 8. 

purvam dnu prdyatim Adade vas tri/i yukta astav aridhayaso 
gah : subdndhavo ye vUy& iva vra d-'tiasvantah srdva aisanta 
pajrah. Ib. 1. 126. 5. 

te manvala> prathamdin nama, dhenos tri.k saptd matuh para- 
mani vindan : tdj jatwtir abhydnwata vra avtrbhuvad arunir 
ya6d$a goh. Ib. 4. I. 16. 

gobhir ydd im anyi asmdn mrgd-in na vrti mrgdyante : abhit- 
fidmnti dhenubhih. Ib. 8. 2. 6. 

xamndrad urmlm ud iyartl Veno nahhojah prsthdm haryatd- 
$ya darsi: rtdsya sfinav ddhi vistdpi blirat samdndm yonim 
ahhydnusata vrah. Ib. 10. 123. 2 

Previous Scholars: Sayana on RV. gives various meanings to this 
word: 'night" or 'dawn' (1. 121. 2), samyaqananahetava apcih, mmariah, 
ta gaccanfiti samnnaga vidywtdb, vra vratah, mdyutsamgha iva (1. 124. 8), 
vriyanta ill vrah vratah iakaralopnx chandasah (1. 126. 5), vru ity usaso 
riama (4. i. 10), and vra varitaro jaladibhir upayair nirundhana vyadhah 
(8. 2. 6). On AV. 2. 2. 1 he has vrah= either avrtatmanah prajah, or, 
v^riyamanah... *apah, and on 11. 7 (0). 3, vrah varako varunah drah 
dravakah amrtamayah sornah. Geldner, Trans, of RV. p. 149. n., summarises 
the opinions of leading moder n scholars and gives his own opinion : " vra 
nach Roth : Haufe, Schar, nach Piachel : Woib, iiach Bergaigne (40. h. 14) 
femelle en rut oder femrae amoureuse. Dies wohl richtig, vra ist das 
Lockweibchen. In 8. 2. 6. bezeichnet es die weiblichen Locktiere, mit denen 
Jager das Wild einf angen wie di o Sanger den Indra. 1 . 124. 8 sind die Frauen 
gemeint, die auf die Hochzeit g ehen urn Manner einzuf angen (vgl. AV, 6. 
60); 1. 126. 5 mit dem Beiwort vUyah die Hetaren (Sij/ioatat TratStaat) ; 
4. 1. 16 ; 10. 123. 2; AV. 2. 1. 1. die Lockstimmen der Sanger, wobei das 
Bild der nach dem Stier brullenden Kiihe festgehalten wird. 1. 121. 2. ist 
das den Biiffel lockende Weibchen, das Indra als dor Biiffel fur sich 
eelbst erzeugt hatte, gemeint, also eine maya". Whitney renders with 
' troops % although he refers to Pischel's interpretation. 

Fischers view (and evidently also that of Bergaigne and 
Oeldner) that vra- means f a woman', Ved. Stud., II. 131 and 
313-322, rests on the following: (1) corresponding to samanaga 



VOL. XVII. PT. I.] INTERPRETATION OF DOUBTFUL WORDS. 9$ 

iva vrah RV. 1. 124. 8. there are RV. 4. 58. 8, 6. 75. 4, 7. 2. 5, 
10. 86. 10, 10. 168. 2, etc., where women are said to goto a samaria, 
which is shown to have been a kind of festive gathering ; (2) in 
RV. 1. 121. 2, like menamd6vasya and matdram goh, vrU- should 
also refer to a female animal, namely, that of the mahisdh ; (3) 
in RV. 8. 2. 6 there may be a reference to capturing,, eleph" ts 
with the help of a she-elephant, vrah standing for vrabhih (he refers 
to Bergaigne, Melanges Renier p. 88ff.) ; (4) in 1. 126, 5 vifyah 
6 belonging to the people (vi6-) ' qualifying vrah may make it 
equivalent in sense to ve&ya- ' courtesan '; and (5) in the phrase 
abhy dnusata vrah the word may stand for the c prayers ', which 
are often compared to women. 

Against this may be considered the following: (1) corres- 
ponding to samanaga iva vrah RV. 1. 12. 48 and visya iva vrah 
ib. 1. 126. 5, we have samanaga aSucaj jatdveddh ib, 7, 9. 4 (cf. 
Pischel, loc. cit., p. 3 15) [and via a kseti vi&yo vi&am ib. 10. 91. 
2, the only other recorded instances of samanaga- and visya-, 
both referring to 'fire'; (2) the meaning of vend-, with which 
vrah appears in AV. 2. 1. 1 and RV. 10. 123. 1 is .uncertain; 
sometimes the word appears as an epithet (medhavin- ' wise ' 
Naigh. 3. 15., ' eager ', c loving ', BR.), at others as a particular 
deity. In the latter case it is particularly associated with the 
sun, cf. RV. 1. 83. 3, tdtah suryo vratapavend ajani, ib. 10. 123. 
and particularly v. 1, ay dm Vend codayat pfSnigarbha jyotir- 
jarayu rdjaso vimane : imam apam sangame suryasya Mum nd 
vipra matibht rihanti, and AV. 2. 1. 1 and 4. 1. 1 which seem 
clearly to refer to the rising of sun and the consequent unravelling 
of the universe, erstwhile immersed in darkness. Apparently on 
these rest the statements of Nir. 1. 7., SBr. 7. 4. 1. 14, etc., that 
it means the ' sun '. The root ven- ' to long for ' seems identical 
with Av. vaen- ' to see ' and in the latter sense vend- may certainly 
refer to the sun as the 'observer'; (3) abhy dnusata may as 
well be the 3rd aor. plur. of us- c to burn ' (cf . us-, usda- * dawn * 
which would be intended by vrah in this case) with abhi and 
dnu ; and (4) mrgdm nd- vra mrgdyante in RV. 8. 2, 6 may refer 
to the chasing of wild beasts with 'firebrands' in hunting, 
reading vra (inst. sing.) instead of vrah or understanding it 
with Pischel as elliptical for vrabhih . 



94 rNTEBPBETATION OF DOUBTFUL WORDS. [j.B.O.B.S. 

The above suggests that ' glow ' either of fire or of dawn, is 
perhaps the true meaning of vrti- and this is also supported by 
the I.-E. element *ur- (or *ul-) ' to burn ', found here as well as 
in the following : ulka- 'firebrand' or * meteor', ulkusi-(mant-) 
and ulmuka- 'firebrand 5 , ulbana- 'hot' Vj. 196, ll,vdrcas- 
( splendour '; old HG. walm ' warm ', old Bulg. vafb ' glow-heat ', 
variti 'to cook', Lith. isz-vora 'soft boiled vegetable food', 
Lett, wars ' soup ', want * to cook ' Leskien, Alt. Bulg. Ele- 
ment, p. 13 old Slav, varu 'heat' vr$ti 'to cook, to be hot', 
vrutft ' hot ', Lith. vlrti 'to cook' Kluge. Etym. Wb. d. deutsch. 
Sprache, s.v. warm etc. Cf. also Walde, s.v. 3. uer- 9 p. 269. 1 

samuspala- 

samvdnam samuspala bdbhru kdlyani sdm mida : dmum 
ca mam ca sdm nuda samandm hrdayam krdh^, 6. 139. 3. 

Previous Scholars : Say ana hazards a conjecture, ftamuspala so/tnyak 
uptaphala sati. Whitney: " A conciliator, a love-avakener (?), do thou, 
O brown, beauteous one, push together ; push together both yon woman 
and me; make [our] heart the same". He remarks, "The mss. hardly 
distinguish sy and sp, but ours, in general, seem as distinctly as the case 
admits, to read samusyalft in a; yet SPP. has -uapa- (noting one ms. 
as reading -usya-), and, as he has living scholars among his authorities, 
the probability is that he is right. Save here and at XIV. 1. 60. (usyalani 
or uspa-), the word appears to he unknown ". 

Samuspaln- is probably identical with * samul-spara- *" con- 
quering ' ( : spr- ' to conquer '), entailing a loss of final unexploded 
t before s as in ut-stha-: *ustha-: Vern. nth-. See Turner, 
Bulletin of the School of Oriental studies, V, p. 130f. Compare 
jdyanli and spdram in 5. 5. 3, vrksdm vrksam ft rohasi vrsan- 
ydtl 9 va kanydla : jdyantl pratyatisthanti spdram nama va asi, 
with samvdnam 'winning over' and samuspala in the above 
verse. 

tJspala- ' framepieces of a bedstead' in 14. 1. 60, Bhdgas 
tataksa cattirah padan Bhdgas tataksa calvary uspalani : Tvdsta 



1 Walde, loc. cit. 9 tries to show that the meaning, in the Baltic and 
Slavonic languages, * to be hot * etc. came from * to cook ' < 'to stir ' < 
'to mix water '< yer- 'to make wet', and suggests that Arm. varem ' I 
light % var * burn ' and Germ, warm are to be separated from them. He 
does not take into account the Sans, examples amongst them. 



L. XVII. PT. I.] INTERPBETAITON OF DOUBTFUL WORDS. 95 



madhyato *nu vdrdhrdn sa no astu sumangali, seems to 
altogether a different word from the above. It can, however, 
rdly be different from utpala- in Kaus. 35. 26, dtrghotpale 
grhya samvi&ati and 36. 7, 6ayanapadam utpale ca, the 
mer of which the commentators explain, mamcakexe (??) 
ob.- kese, the (side-)poles of the bedstead] adhah krta (? -tva) 
'a samviMi (Dar.), and khatvdm adhomukhapattikam grhltvd 
____ svapiti (Kes.). Assuming utpala- to be the correct form, 
> appearance of uspala- can be explained as coming first 
ough a Prakritic tendency to uppala- and then by ms. 
ruption becoming uspala-. Other instances of pp ( r ) 
Bearing as sp ( *n ) are found elsewhere, cf . Whitney's remarks 
<h regard to pippali- at 6. 109. 1 and pippala- at 9. 9. 20. 



langalam pamrMal su&imam sotnasdtsaru : ud id vapatu 
^ dvim prasthavad rathavahanam pivanm ca prapJtarvyam. 
17. 3. 

Previous Scholars : Say ana : ' karsakasya sukJut-karamS BR. suggest 
>ndation to susimam ' having a good furrow '. For somasdtsaru 
somatiat'Saru, VS., MS., KS., and Vas. Dh. -pit-saru, Ppp. -ptisalam), 
ch Whitney considers impracticable, Weber conjectures a noun 
in 'strap* and emends to sorna (=sa-uma) satsaru 'with strap and 
die'. For the rest and varice lectiones see elaborate discussion of 

verse by Whitney, who renders : " Let the plow (itiftgala), lance - 
ited, well-lying, with well-smoothed handle, turn up (itd-vap) cow, 
>p, an ongoing chariot-frame, and a plump wench". 

Su&tma- in SBr. 6. 44. 3, where it is used to explain susad-, 
iently contains the root 3i- c to lie ', see BB. It also appears 
iaus. 76. 23, ise tva sumangali prajdvati prajavati su&rna iti 
thamam, in the following sutra (the same expression), and in 
6, samrchata svapatho 9 navayantah su&imakamav ubhe virdjdv 
e suprajasau, etc. Bloomfield (suggested also by BR.) 
mds in all these cases to susima-, but if su&ila- is substituted 

su&ma- the meaning suits all right. &ima- and Sila- 
,bit ' thus appears to be parallel formations with t- * to lie ', 
Anally meaning simply ' inclination '. In the above verse 
ima- probably refers to the inclination of the pole to the 
mp of the plough. 



96 INTERPRETATION OF DOUBTFUL WORDS. [j.B.O.R.S. 

As most of the verses of this hymn are found also in 
different texts (Ppp., RV., VS., MS., TA., TS., Vas. Dh.) but 
nowhere in the same order or all together, it is possible that 
this verse is culled from a different context and soma may 
well be a vocative and $dtsaru=sdt-tsaru- * well-handled '. 
With this understanding, which does not require change of 
accent or any other supposition, the verse may be translated : 
< Let fche plough, furnished with a lance (i.e., the share), well- 
inclined, of a good handle, Soma, turn up ', etc. 

In this connection may be considered RV. 10. 93. 14, prd 
tad duh$ime pfthavane vene prd rame vocam dsure maghdvateu : 
ye yuktvaya pdnca ata 'smayu patha vi&ravy esam, where both 
duhMma and pfthavana- have been considered proper names. 
Now King Vena, who is notorious in tradition as a perpetrator 
of evil deeds (cf. Sorensen's Index to MBh., s.v. Vena, and 
W. Kirfel, Das Purana Pancalaksana p. 145f ., 15-17 ; 234f . 
10-33; 251, 3-10), had a son called Pythu. Thus, with the 
above meaning ot6ima-, duh&ima- * evil-disposed ' may well be 
his epithet, while pfthavana- may mean 'having Prthu (as a 
son)' see suffix -ana under pdrSana and thus be another 
epithet. 1 

1 The above was approved by the University of London as a thesis 
for the degree of Doctor of Philosophy in July, 1930. 



INDEX OF WORDS. 

Sanskrit. 



U- 26-31. 


6ru- 54-58. 


ya- 31f. 


opaSa- 62-65. 


i-, afljivd- 60. 


kdkutsala- 58f. 


uk$$a- 32-34. 


kandknaka- 59. 


oghavita 34-36. 


Vkam- 64. 


hdlajl- 38. 


kamdla- 59f. 


a- 78-80. 


kalmalf- 60-62. 


sft- 681. 


i/kuftc- 64. 


raparai?ah 36. 


^/kum- 64. 


/ dnusata 93. 


kumara- 64. 


taki- 36-38. 


kumba- 62-65, 


la- 37. 


kumba- 63. 


a- 38. 


kumbha- 64. 


|- 38f. 


kurala- 66. 


a6ayu- 39-41. 


kurira- 62-65. 


*-35. 


kurutin- 66. 


*<* 35. 


kurutini- 65f. 


n- 35. 


kurula- 66. 


rjuiQam 41f. 


komala- 64. 


idanta 74. 


fcowi/ci- 63f . 


>- 90. 


khacjura- 66f. 


ak$abha- 42f. 


khalurika- 66. 


rhsuktagila- 43f. 


khrgala- 67. 


rjata- 44f. 


kfygalya- 67. 


Ilka- 46f. 


gandamala- 68 ^. 


ihih 32, 


gdrda- 56. 


nga- 46f. 


galda- 56f. 


-28. 


galuntal? 67f. 


ava- 47-48. 


ghrtastavas- 68f. 


iradrau 49f. 


citi- 69f . 


ila- 95. 


Jabhya 70f. 


vant- 35. 


tarda- 70f. 


A:vasa 71. 


tayadardm 71-73. 


mit- 29. 


tiritin- 73f. 


a/a- 94f. 


turmi^am 74. 


,ntu 50-54. 


talmatd- 74-75. 


ttra- 52. 


taula- 76. 


($ 52f . 


tativilika- 57. 


yfdha 53. 


^/dika* 33. 



98 



INDEX OF WORDS. 



[J.B.O.R.S. 



duradabhna- 76. 


vra- 91-94. 


duhtima- 96. 


v^^od- 54 1. 


durd~ 77f. 


6aphaka- 46. 


niksana- 33. 


Samara- 60. 


ny- 78-80. 


tarmdri- 61. 


pa turd-, pa{aur&~ 80f. 


6i6ukd' 47. 


pardjvadha- 73. 


sartidamtid- 29. 


pdrasvant- 71-73. 


samankdm 71. 


parimit- 29. 


samupala- 94f. 


pdrisvaftjalya- 29. 


sahasraksd- 30. 


paru$- 81f. 


su^ima- 95f. 


pdrSana- 82f. 


somasdtsaru 96. 


parfoadha- 73. 


hdritau 61. 


paladd- 29. 




pavdsta- 83f. 


Pali and Prakrit. 


patura-, pfyora- 80. 


akkha- (Pr.) 27. 


puruaasya mayu- 81 f. 


akkhavata- (Pa.) 27. 


pfthavana- 96. 


dw&sa- (Pa.) 78. 


pf 9ff- 84f. 


<i?7$a- (Pr.) 78. 


p^ra- 67. 


patthi-, pitthi; put.thi- (Pr.) 85. 


p6tra- 85f. 


palasata-, palasada- (Pa.) 72. 


pratimit- 29. 


2Mfte (Pa.) 85. 


pravartd- 61. 


Vernacular. 


pratarjit- 45. 


(B=Bengali, G=Gujarati, H= 


prartha- 86f. 


Hindi, M=Marathi, N=Nepali, 


6a/^a- 32V 


P=Punjabi, S=Sindhi, T= 


bhrmalsi- 87f. 


Telugu). 


maksumgamd- 47. 


aJfc^aio (N) 30 1. 


man(n)au 88f. 


akhu (B) 27. 


madhyamaSir 89f. 


aJfc^wn (B) 28. 


mayti- 82. 


ukh (H) 28 1, 


mdhina 30. 


gad (B, H) 57. 


mulati- 46. 


^ofca (B) 69. 


mura- 90f. 


dhusa (N) 78. 


muradeva- 91. 


dhussa (P) 78. 


ruksa- 33. 


d^7a (B) 78. 


rdgfa- 48. 


pith (H), pwM (G, M), pwffcS (S) 85. 


vdgf^a- 70f. 


bag(gh)di (S) 70. 


i/van- 48 l . 


baghdulu (B) 70. 


va^^ra-, -A;y^- 48 l . 


wand (P) 89 1. 


vend- 93. 


warm (T) 89 1. 


vetanta- 68. 


mana (B) 89. 



INDEX OF VEDIG PASSAGES. 



AV. 



I. 


2. 4 at 47. 


120. 2 at 40. 




7. 2 at 76. 


138. 1, 2, 3 at 62. 




28. 3 at 90. 


139. 3 at 94. 


II. 


1. 1 at 91. 


VII. 50(52). 2 at 41. 




3. 1 at 35. 


95(100). 2 at 35. 




3. 2, 3 at 47. 


VIII. 2. 16 at 32. 




9. 4 at 69. 


3. 2 at 91. 


[II. 


9. 3 at 67. 


3. 13 at 91. 




14, 1 at 44. 


4. 5 at 82. 




14. 2 at 45. 


4. 24 at 50, 91. 




17. 3 at 95. 


6. 7 at 73. 


IV. 


7. 6 at 77, 83. 


6. 9 at 59. 




9. 1 at 31. 


6. 11 at 77. 




9, 4 at 89. 


8. 18 at 26. 




9. 8 at 32. 


IX. 2. 22 at 70. 




12. 2 at 85. 


3. 8 at 26. 




15. 9 at 35. 


8. 20 at 38. 




17. 3 at 90. 


X. 1. 15 at 65. 




34. 5 at 45. 


4, 2 at 81. 




36. 9 at 39. 


4. 22 at 59. 




37. 6 at 36. 


XL 1.2 at 34. 


V. 


13. 6 at 74. 


2. 30 at 43. 




17. 16 at 45. 


4(6). 23 at 79. 




18. 4 at 75. 


7(9). 3 at 92. 




22. 9 at 86. 


7(9). 4 at 78. 




28. 12 at 44. 


9(11). 14 at 80. 




31. 2 at 64. 


9(11). 16 at 66. 


VI. 


14. 3 at 46. 


XII. 1.4 at 79. 




16. 3 at 67. 


1. 46 at 87. 




22. 3 at 54. 


2. 16 at 79. 




30. 1 at 88. 


2. 17 at 68. 




44. 1 at 48. 


4. 4, 19 at 76. 




44. 2 at 48. 


5. 45, 46 (XII. 11. 7, 8) at 




49. 1 at 50. 


36. 




49. 2 at 49, 54. 


XIV. 1.8 at 62. 




60. 1, 2, 3 at 70. 


XV. 2. 1 (6) at 60. 




65. 1 at 54. 


2. 2 (14), 3 (19), 4 (25) at 




72. 2, 3 at 71. 


61. 




83. 3 at 67. 


XVIII. 4. 66 at 58. 



100 



INDEX OF VBDIO PASSAGES. 



[j.B.O.B.S. 



XIX, 6. 4 at 79. 

7. 1 at 74. 

49. 1 at 42. 

49. 4 at 82. 
XX. 135. 11 at 61. 

.KS. 

V, 7. 11 at 71. 
13. 11, 12 at 80. 

MS. 

II. 7. 12 at 67 

RV. 

I. 121. 2 at 92. 
124. 8 at 92. 
126. 5 at 92. 

133. 6 at 42. 

134. 6 at 41. 
171. 3 at 63. 
180. 5 at 27. 

II. 24. 3 at 74. 
33. 8 at 61. 
39. 4 at 67. 

III. 29. 9 at 34. 

IV. 1. 16 at 92. 
17. 2 at 53. 

VII. 104. 5 at 82. 

104. 24 at 50. 
VIII. 1.10 at 79. 

1. 20 at 57. 

2. 6 at 92. 
7. 34 at 82. 



22. 10 at 47. 
27. 11 at 79. 
45. 41 at 82. 
72. 2 at 88. 
77. 11 at 52. 
104. 24 at 91. 
IX. 63. 8 at 88. 
65. 16 at 88. 
X. 27. 7 at 83. 
85. 8 at 62. 
*6. 18 at 71. 
87.2at9J. 
87. H at 91. 
97. 12 at 89, 
123. 2 at 92. 

ankh. Aran. 
3. 2 at 72. 

SV. 
I. 4, 1. 2. 5 at 57. 

TS. 

III. 1.1 1,8 at 54. 

IV, 1. 5,'3 at 62. 
V, 5. 21 at 71. 

6. 21, 1 at 37. 

7. 21-22 at 80. 



XI. 

XII. 

XXIII. 

XXIV. 



VS. 

56 at 62. 
86 at 89. 
21 at 60. 
28 at 71.